Grid-connected solar PV farms are promising ways to foster the development and utilization of renewable energy resources. However, power grids with high penetration of renewable energy sources are characterized by low system inertia and are susceptible to frequency and voltage instability. The Grid Forming (GFM) control strategy can mimic the inertia response characteristics of a synchronous generator by providing grid support. However, the conventional droop-based grid-forming control suffers from a fixed droop gain problem, rendering it incapable of delivering adequate damping to mitigate post-fault instability. Thus, this paper proposes a deep reinforcement learning (DRL)- based adaptive droop control for simultaneously supporting frequency and DC voltage in grid-connected solar farms. The proposed method interacts with the environment to learn the optimal droop gain for optimal operation, frequency regulation, and DC bus voltage stabilization. The effectiveness of the proposed method is validated in MATLAB/Simulink. The comparison results with the conventional method confirm the superior performance of the proposed method under load variation and fault conditions.

With the acceleration of urbanization, climate change, and population growth, the global electricity demand shows a significant upward trend. Accurate short-term load forecasting (STLF) plays a vital role in optimizing the operation of the electrical distribution system. Although recent deep learning-based short-term load forecasting models have shown significant advantages, achieving accurate load forecasting remains a daunting challenge as power load demand is affected by many external environmental factors and the inherent defects of traditional forecasting models such as recurrent neural networks (RNNs) and support vector machine (SVM). In order to tackle this challenge, this paper proposes a transformer-based short-term load forecasting model. It takes loads in distributed feeders as forecasting objects and makes full use of the self-attention mechanism to capture the long-term dependency and complex nonlinear characteristics of load data. Experimental results show that the model performs well in processing complex time-series data and load fluctuations in different seasons. It has strong generalization ability and provides a new solution for forecasting distribution feeder load.

This paper proposed a physics-constrained twin delayed deep deterministic policy gradient (TD3) algorithm for simultaneous virtual inertia and damping control of a grid-connected variable speed doubly-fed induction generator (DFIG) wind turbine using a combined deep reinforcement learning (DRL) and quadratic programming as a novel solution to suppress frequency fluctuations caused by the control mechanism which decouples the active power from the system frequency, thus hiding the rotating kinetic energy of the wind generator. The optimization stage modifies the action of the DRL agent, thus preventing the agent from taking certain unsafe actions. We tested the effectiveness of the proposed scheme under various scenarios through simulations on an IEEE 9-bus test system in MATLAB/Simulink. Compared with other virtual inertia controls, the results show that the proposed scheme achieved improved dynamic performance with the lowest system frequency deviation and fastest frequency recovery under wind and load variations and severe grid faults. A further test on the IEEE 39-bus system shows that the grid size does not affect the performance of our proposed technique.Note to Practitioners—Integrating the wind turbine systems into the utility grid results in power quality problems such as frequency fluctuation, voltage dip, power loss, and severe power outages. The unpredictability and uncontrollability of the wind pose a serious problem in integrating wind energy conversion systems. This problem becomes worse with the increasing number of connected wind turbines. Therefore, new control strategies are required to mitigate this issue. The droop-based virtual and damping control method traditionally provides frequency support in grid-tied wind turbine systems. However, the fixed droop gain is a significant drawback of this method. In this paper, we proposed a novel physics-constrained deep-reinforcement learning-based virtual inertial and damping control. The proposed control agent is constrained from unsafe actions and rewarded for maintaining the grid frequency within operational limits. Simulation results with the IEEE 9 bus system validated the feasibility and effectiveness of our proposed approach. A comparison of our method with the conventional control scheme, adaptive droop-based virtual inertia control, etc., carried out under various operational scenarios verified the enhanced performance of our proposed strategy.

The coexistence of high-frequency phasor measurement units (PMUs) and conventional SCADA systems raises the challenge of heterogenous-source and multi-rate measurements which significantly degrades the performance of power system dynamic state estimation. In this work, a deep reinforcement learning (DRL) assisted robust cubature Kalman filtering (CKF) scheme is proposed to handle measurements from hybrid sources and with different time scales. In specific, a multi-rate measurement function reconstruction approach is designed with an independent discretization mechanism to lift the present limitation of requiring an integer multiple relationship of the sampling rates from multiple sources in most of existing works. Embedded with this discretization mechanism, a deep reinforcement learning assisted two-parameter linear exponential smoothing method is proposed to reconstruct the slow measurement model with online adjustable estimation parameters. A generalized correntropy loss criterion is also included in the robust CKF to counter the non-Gaussian noise and the noise distribution variation caused by the reconstruction. Comparisons results demonstrate that the proposed DRL-based robust CKF method can achieve better accuracy and robustness under various operating scenarios.

This article studies the issue of stability in memristor-based neural network (MNN) systems with time-varying delays. First, a novel matrix-separation Legendre inequality is proposed to achieve a tight hierarchical bound on augmented-type integral terms. To derive implementable inequality conditions, several delay-dependent matrices are introduced to eliminate the reciprocal terms associated with time-varying delay. Furthermore, a new Lyapunov-Krasovskii (L-K) functional is proposed by incorporating augmented-type double integrals and delay-product terms. A series of free-weighting matrices are introduced into the L-K functional, leveraging the zero-sum equations and the S-procedure pertaining to both the delay and its derivative. Based on the proposed matrix-separation Legendre inequality and L-K functional, the derived stability conditions exhibit reduced conservatism, as validated by three numerical cases and simulation results.

This work develops a distributed proportional-integral (PI) controller combined with the signal-to-interference-plus-noise ratio (SINR)-dependent dynamic event-triggered (DET) communication strategy to cope with the frequency regulation and power sharing problems subject to jamming attacks in an inverter-based islanded microgrid. In order to simplify the controller structure, power sharing and frequency restoration can be implemented simultaneously in the designed secondary control layer instead of the traditional hierarchical control implementation. Besides, the dynamic event-triggered communication strategy configured with each secondary controller can adaptively adjust the triggered threshold based on the SINR signal, which can reduce the congestion of communication networks caused by jamming attacks and ensure system control performance. A cyber-physical microgrid testbed is built based on the real-time simulator, OPAL-RT, and network simulation software, EXata, to verify the effectiveness of the proposed SINR-dependent event-triggered based distributed secondary controller.

Although several multi-agent deep reinforcement learning (MADRL) algorithms have been employed in power distribution networks configured with high penetration level of Photovoltaic (PV) generators for active voltage control (AVC), the impact of the voltage fluctuation of a single PV node on voltage violations of other PV nodes in the network is ignored. Consequently, it leads to the conservativeness of the existing MADRL based AVC algorithms. In this paper, a robust MADRL control algorithm is designed to minimize the nodal voltage violation and line loss with the exploration of coupling voltage fluctuations across all the controlled nodes by coordinating PV inverters, and a physics factor is utilized to guide (physics-guided) the training policy with the expectation of a better performance compared to existing purely data-driven methods. In the proposed physics-guided multi-agent adversarial twin delayed deep deterministic (PG-MA2TD3) policy gradient algorithm, a physics factor, global sensitivity of voltage (GSV), is properly embedded in the algorithm to measure the influence of the nodal voltage fluctuation on voltage violations on the other controlled nodes with PV inverters and this GSV is shared in the learning center to guide the centralized learning and decentralized execution process. The multi-agent adversarial learning (MAAL) embedded with the GSV to seek an adaptive descend gradient for reducing the Q-value function appropriately rather than always assuming the worst case. Therefore, this physics-guided method can reduce the conservation and provide significantly better reward. Finally, the proposed algorithm is compared with several other methods on IEEE 33-bus, 141-bus and 322-bus with three-year data in Portuguese and the results indicate the proposed method can obtain the minimal voltage fluctuation and the best reward in the comparisons.

To defend the cyber–physical system (CPSs) from cyber-attacks, this work proposes an unified intrusion detection mechanism which is capable to fast hunt various types of attacks. Focusing on securing the data transmission, a novel dynamic data encryption scheme is developed and historical system data is used to dynamically update a secret key involved in the encryption. The core idea of the dynamic data encryption scheme is to establish a dynamic relationship between original data, secret key, ciphertext and its decrypted value, and in particular, this dynamic relationship will be destroyed once an attack occurs, which can be used to detect attacks. Then, based on dynamic data encryption, a unified fast attack detection method is proposed to detect different attacks, including replay, false data injection (FDI), zero-dynamics, and setpoint attacks. Extensive comparison studies are conducted by using the power system and flight control system. It is verified that the proposed method can immediately trigger the alarm as soon as attacks are launched while the conventional$\chi^{2}$detection could only capture the attacks after the estimation residual goes over the predetermined threshold. Furthermore, the proposed method does not degrade the system performance. Last but not the least, the proposed dynamic encryption scheme turns to normal operation mode as the attacks stop.

In this article, a novel distributed fuzzy load frequency control (LFC) approach is investigated for multiarea power systems under cross-layer attacks. The nonlinear factors existing in turbine dynamics and governor dynamics as well as the uncertain parameters therein are modeled and analyzed under the interval type-2 (IT2) Takagi–Sugeno (T–S) fuzzy framework. The cross-layer attacks threatening the stability of power systems are considered and modeled as an independent Bernoulli process, including denial-of-service (DoS) attacks in the cyber layer and phasor measurement unit (PMU) attacks in the physical layer. By using the Lyapunov theory, an area-dependent Lyapunov function is proposed and the sufficient conditions guaranteeing the system’s asymptotically stability with the area control error (ACE) signals satisfying$\mathcal {H}_{\infty }$performance are deduced. In simulations, we adopt a four-area power system to verify the resiliency enhancement of the presented distributed fuzzy control strategy against random cross-layer DoS attacks. Results show that the designed resilient controller can effectively regulate the load frequency under different cross-layer DoS attack probabilities.

In order to ensure that the power system frequency and tie-line power remain at the nominal value when the load fluctuates, while reducing the release number of decentralized sensor, this work presents a novel adaptive event-triggered scheme for the load frequency regulation via designing the decentralized dynamic output feedback controller (DOFC), where the communication delay issue is also considered due to communication constraints. Distinct from the existing ones, the proposed adaptive event-triggered strategy automatically tunes the threshold according to the local extremum of the system output signal, which can significantly reduce the number of unnecessary signal transmissions to ensure system performance. First, the proposed adaptive event-triggered transmission scheme is integrated with the decentralized DOFC and communication delays under the framework of a linear time-delay system. Then, the asymptotic stability of the closed-loop power system is analyzed through the Lyapunov stability theory and a procedure is given for the design of decentralized dynamic output load frequency controllers by solving some linear matrix inequalities (LMIs). Finally, a three-area power system is used to verify the effectiveness and usefulness of the proposed results.

With the participation of a significant number of plug-in electric vehicles (PEVs), it is really challenging to achieve economic-effective in load frequency control (LFC) while sustaining satisfiable system performance. To tackle this challenge, a new distributed economic model predictive control (DEMPC) strategy is proposed for the LFC with the large-scale PEV participation. In the light of the vulnerability of LFC to false data injection (FDI) attacks, a model-based χ2intrusion detection unit is integrated with the proposed DEMPC. This model-based intrusion detection unit can not only monitor the FDI attacks, but also generate a model-based state prediction for the DEMPC once the data is identified as compromised. Then, an event-triggering mechanism is presented to reduce the computation and communication burdens of each area controller. Simulation studies of a four-area power system are conducted and the results validate the effectiveness of the proposed intrusion detection unit and event-triggering conditions for the DEMPC.

While communication-based train control (CBTC) systems play a crucial role in the efficient and reliable operation of urban rail transits, its high penetration level of communication networks opens doors to Man-in-the-Middle (MitM) attacks. Current researches regarding MitM attacks do not consider the characteristics of CBTC systems. Particularly, the limited computing capability of the on-board computers prevents the direct implementation of most existing intrusion detection and defense algorithms against the MitM attack. In order to tackle this dilemma, in this article, we first introduce edge intelligence (EI) into CBTC systems to enhance the computing capability of the system. A cross-layer defense scheme, which includes the detection and defense stages, are proposed next. For the cross-layer detection stage, we propose a Long Short-Term Memory (LSTM) and Support Vector Machine (SVM) based detection method to combine the detection probability calculated from the train control parameter sequence and operation log files. For the cross-layer defense stage, we construct a Bayesian game based defense model to derive the optimal defense policy against MitM attacks. To further improve the accuracy of the defense scheme as well as optimize the communication resource allocation scheme, we propose an optimal communication resource allocation scheme based on the Asynchronous Advantage Actor-Critic (A3C) algorithm at last. Extensive simulation results show that the proposed scheme achieves excellent performance in defending against MitM attacks.

In this paper, one new distributed load frequency regulation approach is proposed for smart power system operation under two specific practical constraints, including the limited communication resource and speed droop parametric uncertainty. To address these two constraints, the co-design of event-triggering communication scheme and distributed model-based controller is studied. Instead of using zero-order holders, the proposed model-based scheme is able to extend the maximum allowable time interval and thus reduce communication bandwidth usage. In the meantime, the proposed co-design scheme is able to get the model-based control parameters and event-triggering condition metrics simultaneously. This can loosen the conservation in the choice of control gains and event-triggering parameters faced by existing approaches where the control gains are fixed in prior. Comparisons on the multiple-area system confirm that this designed load frequency regulation method significantly reduces the number of required data transmissions without sacrificing the dynamic performance of the frequency and tie-line power. It is also shown that the proposed approach has great robustness to speed droop coefficient uncertainty.

As communication networks are increasingly implemented to support the information exchange between microgrid control centers and/or local controllers, they expose microgrids to cyber-attack threats. This paper aims to analyze the stochastic stability of islanded microgrids in the presence of random denial of service (DoS) attack and propose a mode-dependent resilient controller to mitigate the influence of DoS attacks. Specifically, the small-signal model of the microgrid under the DoS attack is integrated as a stochastic jump system with state continuity disruptions. A new vulnerability metric is defined by using observability Gramians of the stochastic jump system, to measure the vulnerability of the system regarding DoS attack choices. The Lyapunov function analysis is conducted to find conditions sustaining the stochastic stability of the islanded microgrid in the form of linear matrix inequalities. A mode-dependent control approach is proposed for microgrids to mitigate the influence of random DoS attacks. In case studies, the vulnerability analysis and time-domain simulation results show the performance of the investigated microgrid can be degraded when the random DoS attacks exist. When the proposed mode-based secondary frequency controllers are installed, the islanded microgrid can sustain its stability during the attacking period and system dynamics rapidly converge when the DoS attack is over.

An integrated model-based control and scheduling scheme is proposed for the load frequency control (LFC) of large-scale power systems under the distributed structure and uncertainties. Specifically, the limited bandwidth constraint is considered when state observation is exchanged over shared communication networks. Each area controller uses the explicit models of its own and neighboring areas to predict state observations when the actual one is not available. At each transmission instant, the state observation of the scheduled area is broadcasted to the relevant areas and the model-based controllers are partially updated. By properly scheduling the transmission sequence and intervals, the stability of the power system can be guaranteed with a substantial reduction of the bandwidth usage and this is proven by performing a thorough theoretical analysis. Simulation results of a four-area power system verify that the proposed distributed model-based control scheme integrated with a proper scheduling strategy can greatly enhance the performance and the resiliency to parameter uncertainty in large-scale power systems.

This paper investigates the uncertainty of the day-ahead distribution system scheduling considering the random variations of both Photovoltaic-based distributed generator (PV-DG) output power and load. Instead of Monte-Carlo simulation (MCS), a two-point estimation method (2PEM) is applied to obtain accurate and computation-efficient analysis results. Based on the two-year real-world hourly weather and load data in the city of Ottawa, the estimation accuracy of the 2PEM has been verified in an equivalent 44 kV distribution feeder system. In terms of computational efficiency, the 2PEM can significantly reduce the computation burden with comparison to MCS. By using the 2PEM, the impact of PV-DG output power and load variations on the uncertainty of the distribution system scheduling under different seasons is thoroughly studied. The analytical results show that the range of standard deviation of optimally scheduled DG generation for this distribution feeder system is larger in summer than that in winter.

This paper considers the master and slave synchronization control for bilateral teleoperation systems with time delay and backlash-like hysteresis. Based on radial basis functions neural networks' approximation capabilities, two improved adaptive neural control approaches are developed. By Lyapunov stability analysis, the position and velocity tracking errors are guaranteed to converge to a small neighborhood of the origin. The contributions of this paper can be summarized as follows: 1) by using the matrix norm established using the weight vector of neural networks as the estimated parameters, two novel control schemes are developed and 2) the hysteresis inverse is not required in the proposed controllers. The simulations are performed, and the results show the effectiveness of the proposed method.

This work investigates the impact of the communication-channel cyber attacks on the dynamic performance of the islanded microgrid secondary frequency control. The cyber-physical system structure of the secondary frequency control is described. The secondary frequency control is introduced. A set of cyber attacks including denial of service (DoS) and false data attacks are then modeled. The Canadian urban benchmark distribution system has been built for testing the impact of cyber attacks. The testing results show that both DoS and false data attacks could result in the performance degradation and even the instable operation of the islanded microgrid.

Due to its great potential to improve the overall performance of data transmission with its dynamic and adaptive spectrum allocation capability in comparison with many other networking technologies, cognitive radio (CR) networking technology has been increasingly employed in networking and communication infrastructures for smart grids. However, a secondary user (SU) of a CR network has to be squeezed out from a channel when a primary user reclaims the channel, which may occur in a randomized fashion. The random interruption of SU traffic may cause packet losses and delays for SU data, and it will in turn affect the stability of the monitoring and control of smart grids. In this paper, we address this problem and investigate the modeling and stability analysis of the automatic generation control (AGC) of a smart grid for which CR networks are used as the infrastructure for the aggregation and communication of both system-wide information and local measurement data. For this purpose, a randomly switched power system model is proposed for the AGC of the smart grid. By modeling the CR network as an On–Off switch with sojourn times, the stability of the AGC of the smart grid is analyzed. In particular, we investigate the smart grid with two main types of CR networks: 1) the sojourn times are arbitrary but bounded and 2) the sojourn times follow an independent and identical distribution process. The sufficient conditions are obtained for the stability of the AGC of the smart grid with these two CR networks, respectively. Simulation results show the effects of the CR networks on the dynamic performance of the AGC of the smart grid and illustrate the usefulness of the developed sufficient conditions in the design of CR networks in order to ensure the stability of the AGC of the smart grid.