The integrity of distribution network state estimation is critically challenged by false data injection attacks, whose detection is often hampered by the numerical instability of underlying estimators. Such instability introduces artifacts that can mask an attack’s signature. This article presents a unified framework that achieves robust detection by decoupling these estimator artifacts from malicious data patterns. The framework’s core is a numerically stabilized forecasting aided state estimation employing a U-D factorization cubature Kalman filter (UD-CKF). By ensuring covariance positive-definiteness, it generates high-fidelity state estimates and, crucially, a statistically consistent innovation covariance matrix. This stable foundation enables a novel geometric inconsistency detector (GID). Instead of analyzing temporal patterns, the GID evaluates the geometric alignment between the observed innovation vectors and their expected statistical distribution defined by the UD-CKF. By monitoring the evolution of the innovation subspace, it effectively distinguishes the random orientation of noise from the persistent directional signature of a stealthy attack. This approach is validated on IEEE test feeders, demonstrating exceptional capability in identifying subtle FDIAs that remain undetected by magnitude-based or purely temporal methods.

Porcelain insulators are prone to degradation, including internal defects and mechanical fractures, due to long-term exposure to harsh operational stresses. Such failures threaten the safe and reliable operation of power grids. Existing online monitoring techniques often suffer from limitations in accuracy, safety, cost, or scalability. To address these challenges, this article proposes a novel online monitoring and intelligent fault diagnosis system for porcelain insulators. A low-cost, nonintrusive device is designed to measure voltage distribution across insulator strings and transmit data wirelessly. An intelligent diagnostic algorithm based on voltage patterns is then developed to assess insulator health, issue early warnings, and pinpoint fault locations. Simulation results confirm the algorithm’s predictive accuracy, achieving a mean squared error of 0.028–0.051 and a mean absolute error of 0.058–0.098 across multiple voltage levels and environmental conditions. More critically, field deployment on live transmission lines demonstrates a fault prediction accuracy exceeding 99.5% during two months of continuous operation. These results validate both the robustness and scalability of the proposed method. This article thus provides a practical solution for predictive maintenance, offering real-time, high-precision diagnostics that significantly enhance the operational intelligence and safety of modern power grids.

Managing energy usage flexibility has been identified as an effective way to coordinate Internet of Things (IoT) technologies and transition to a low-carbon future. However, the various models of these flexibilities may not be fully understood, and some may even be characterized by closed-box neural networks, such as those used for electric vehicles (EVs) charging. In this article, we propose a novel framework called augmented shadow-price deep reinforcement learning (ASP-DRL) for the online, distributed management of IoT under multiple sources of uncertainties, including renewable energy sources (RESs), wholesale electricity prices, and EV behavior patterns. To address these challenges, the proposed framework combines scheduling mechanisms for neural networks and optimization models to maximize total social welfare. Within the ASP-DRL framework, the policy network adaptively learns about system uncertainties and delivers actions to different distributed entities, either to form augmented objective functions for optimization models or to guide neural networks. We also present a distribution correction algorithm that enhances the vanilla soft Actor-Critic method with attention-based maximal corrective feedback, resulting in faster convergence and better performance. Our numerical studies demonstrate the superiority of the proposed ASP-DRL framework compared to conventional deep reinforcement learning (DRL) and optimization-based approaches.

The concept of the virtual synchronous generator (VSG) is recognized as a crucial technology for addressing the challenges posed by the low inertia of power systems. However, in practical microgrids, the inertia parameters of VSGs may differ, fluctuate, or be unknown, and the inertia at their connection points can be affected by other devices. These uncertainties complicate the analysis of inertia characteristics and the development of compatible controllers. To address these issues, this article proposes a distributed secondary adaptive frequency coordination and oscillation suppression control method for heterogeneous VSGs in islanded microgrids. By incorporating an adaptive control parameter based on the rate of change of frequency and consensus tracking errors in the distributed control, our method can be conducted without relying on predetermined inertia parameters and can ensure consistent suppression of frequency oscillations. In addition, the permissible stable boundaries of the control parameters in the presence of delays are examined using a Lyapunov–Krasovskii candidate. The effectiveness of the proposed method is ultimately validated through simulation cases and hardware-in-the-loop experiments.

A phasor measurement unit (PMU) serves as a superior tool to monitor the dynamics of the power system, but its high cost remains a practical concern that requires the optimal placement of the PMU (OPP). Traditionally, researchers relied on model-based approaches to analyze this problem. However, these methods not only suffer from inevitable parameter uncertainties but can also be computationally expensive for complicated power system dynamic models. Faced with these issues, this article proposes a data-driven OPP approach utilizing an augmented Koopman operator. This operator lifts the original nonlinear state space to a high-dimensional linear Koopman space in a data-driven manner, which fully eliminates the model discrepancy while achieving high computing efficiency. Theoretically, we prove that the observability matrix in the augmented Koopman canonical coordinates preserves the whole dynamic evolution of both the system model and its associated measurement model. Finally, we propose a modified genetic algorithm to solve the established OPP problem, which is enhanced to further accelerate the search speed. The simulation results reveal the excellent performance of our proposed method.

Power distribution networks are evolving toward a modernized cyber-physical system that is vulnerable to mounting cybersecurity threats brought by false data injection attacks (FDIAs). To address the issue, this article innovatively proposes a multiarea probabilistic forecasting-aided interval state estimation (MPF-ISE) framework for FDIA identification in power distribution networks. The framework devises a novel probabilistic forecasting (PF) approach to achieve interval pseudo measurement modeling. A nonlinear programming-based training algorithm is formulated to minimize the interval width of forecasting error quantiles, which are analytically approximated by Cornish-Fisher expansion. Next, real-time interval measurements are constructed based on the unknown-but-bounded theory. By implementing interval measurement conversion, the local PF-ISE model is converted into rectangular coordinate forms and iteratively solved using a modified Krawczyk operator. Then, the local model is extended to a multiarea form and solved by considering the information exchange between adjacent subareas. Finally, the solution of MPF-ISE, which takes into account measurement uncertainties and line parameter variations, is regarded as the normal operating level, thus forming the proposed FDIA identification scheme. Case studies on the modified IEEE 123-node test feeder demonstrate the superiority of the proposed PF and the effectiveness of MPF-ISE in identifying FDIAs compared with existing ones.

The state estimation method is troubled by heavy computational tasks and poor estimation tracking capability for the large-scale active distribution network. Given the aforementioned difficulty, in this article, we proposed a novel multiarea forecasting-aided state estimation (FASE) strategy to perceive the state of the system effectively. The proposed strategy begins with the implementation of an improved multiarea FASE model. The processing of multisource measurement data, such as microphasor measurement units and supervisory control and data acquisition, and equivalent load-based information interaction reliably complete the FASE of multiareas. Especially, a third degree dimensionality reduction square root cubature Kalman filter (SR-CKF) algorithm is designed for local FASE model considering the influence of large-scale distribution networks data on the numerical stability of the estimator. The case study shows the advantages of the proposed strategy in estimation accuracy, efficiency, and numerical stability compared with the existing ones.

Widespread advanced metering infrastructure and wide-area monitoring systems generate a significant amount of electricity load consumption data, which can facilitate eliciting end users’ temperature flexibility for demand response programs. However, the direct delivery of users’ load profiles is a threat to users’ privacy. So this paper proposes a privacy-preserving hybrid cloud framework for TCL-based demand response programs, composed of user private clouds and aggregation cloud. User clouds store users’ load profiles and elicit temperature flexibility by the proposed stable temperature-related regression model. In the aggregation cloud, this paper proposes the slope-priority flexibility aggregation method for the mean-variance analysis of aggregate flexibility and the XGBoost-accelerated disaggregation model for real-time selecting users based on users’ fitting coefficients. Hybrid cloud achieves privacy-preserving by separating flexibility eliciting models and aggregation/disaggregation methods into user private clouds and aggregation cloud. Numerical experiments verify that: 1) in user clouds, the stable regression model achieves less predict errors; 2) in aggregation cloud, the slope-priority method can achieve higher aggregate flexibility, and XGBoost-accelerated disaggregating reduces the solving time by nearly three orders of magnitude.

A grand challenge for the state estimation (SE) method in large-scale distribution networks lies in how to deal with the increasing computational tasks. This article addresses the issue and proposes a novel multiarea architecture for the unbalanced distribution network SE method. The first step of the method is to introduce an innovative multiarea state estimation (MASE) model using microphasor measurement units (μPMU) mixed with conventional supervisory control and data acquisition (SCADA) systems, with both the coordinate tensions and synchronization issues considered. The proposed model contains a SCADA measurement delay estimator and a MASE algorithm. Then, the hybrid state estimation (HSE) model is solved in a distributed way. In each subarea, the HSE problem is solved locally with minimal data exchanges among neighbor subareas. Case studies show the accuracy and efficiency enhancements obtainable of the proposed MASE method with respect to existing ones.

Optimal iterative learning control (OILC) has been recognized as an excellent model-based means for regulating batch process with abundant successful applications reported in the past decades but also received considerable criticisms for its poor robustness against model mismatch that is common for many industrial situations. Despite numerous attempts to address the issue, many of them are still not able to yield satisfactory control performance particularly in the presence of a possible combination of time-varying uncertainties and conservatively designed controllers, which may compromise the learning mechanism, hence rendering the robustness issue of OILC far from well explored. This article intends to investigate the aforementioned issue by proposing a new OILC method resting upon the minimization of a dynamic upper bound on tracking error which is distilled from better exploitation of the time variation of uncertainties. We also show that the problem can be formulated in the framework of convex–concave game that can be efficiently solved by a subgradient method with an excellent balance of optimality and computation time. Such a formulation enables us to gain: 1) guaranteed monotonic convergence on tracking error; 2) remarkably reduced conservatism on controller synthesis; and 3) controllable computation complexity. It is further shown that the proposed method is capable of handling nonlinearity, for example Volterra system, a classic representation of nonlinear process. The efficacy of the method is verified by numerical experiments on a continuous stirred tank reactor model.

The increasing penetration of photovoltaic (PV) generators has led to a reduction in the effectiveness of existing strategies for restoring the power distribution network. This article proposes a dynamic robust restoration (DRR) framework for the recovery of outage power considering uncertain PV outputs and demands. This framework is presented in two subsequent steps. In the first step, optimal decisions regarding the network configurations are generated. The second step then computes the modified dynamic Distflow equations and constraints under consideration of the worst operating conditions over the associated uncertainty sets with the aim of maximizing the recovery of outage power. The DRR model is formulated as a bilevel mixed-integer linear programming problem. A decomposition algorithm in a master-sub structure is used to solve the resulting system. The results of case studies show that the proposed DRR model yields obvious advantages over the existing deterministic dynamic restoration model in terms of robustness against system uncertainties.