Fast insecurity early warning is the key technique to resist the dynamic insecurity risk, which becomes intractable due to the strong nonlinearity of hybrid AC/DC grids and the high uncertainty of wind generation. Considering dynamic security constraints and wind power uncertainty, this paper presents an insecurity early warning method based on decision tree (DT) and semi-supervised deep learning. First, semi-supervised deep learning is deployed to estimate the dynamic security limit of the critical interface of hybrid AC/DC grids. The system security is assessed by comparing the actual power transfer of the critical interface with the security limit. Then, operating conditions (OCs) are ranked into different insecure levels according to the type of preventive control actions that is needed to ensure the system security. Finally, oblique DT is utilized to identify insecurity classification boundaries in the wind power injection space. Insecure OC sets are constructed based on these classification boundaries. Simulation results of the real-life Jiangsu-Shanghai interconnected grid in east China demonstrate that the proposed method can fast construct the insecure OC sets corresponding to different insecure levels.

Security early warning is crucial for resisting power system risk. To deal with renewable uncertainty, offline trained early-warning models need to be evolved incrementally. Catastrophic forgetting is main obstacle of model evolution. With the evolution times of early-warning models increasing, the capability of existing methods to mitigate catastrophic forgetting gradually decreases. Knowledge distillation can transfer the learned knowledge from the previous model to the updated model. To mitigate catastrophic forgetting, an incremental model evolution method for security early warning based on knowledge distillation and active learning is proposed. First, an incremental style-based generative adversarial network is formulated to generate renewable power scenarios, which utilizes knowledge distillation to retain previously learned knowledge. An improved concept drift detection method is proposed to determine model evolution moment. Then, an incremental deep regression model based on stacked target-related denoising autoencoder and knowledge distillation is constructed to assess system security index. Finally, active learning is deployed to select informative new samples and solve the objective conflict problem of knowledge distillation. Simulation results of a provincial grid in China demonstrate that proposed method can constantly improve early warning accuracy and effectively retain knowledge learned by previous samples. The proposed method can improve the adaptability of early-warning models to actual power system operating conditions, which enhances the intelligence level of security analysis to copy with renewable uncertainty.

Research on the detection and identification of anomalies in electric power systems is crucial for ensuring their secure and stable operation. Anomaly detection models based on Support Vector Machines (SVMs) effectively process high-dimensional data while maintaining strong generalization capabilities. However, the performance of SVMs significantly depends on the choice of parameters, where improper parameter settings can lead to overfitting or underfitting, consequently decreasing the accuracy of anomaly detection. Furthermore, the dimensions of anomaly data in electric power systems are often unknown, making it difficult for existing methods to maintain a high precision in multidimensional data detection, and the segmentation of such data lacks intuitive display. In response, this article proposes an improved SVM model for electric power system anomaly detection, enhanced by parameter optimization algorithms, alongside a method for nonlinear dimension reduction and visualization using t-Distributed Stochastic Neighbor Embedding (t-SNE). Initially, traditional SVM parameters are optimized using the following four algorithms: Grid Search (GS), Particle Swarm Optimization (PSO), Genetic Algorithms (GAs), and Artificial Bee Colony (ABC) algorithms, in order to establish the optimized SVM model for electric power system anomaly detection. Finally, the effectiveness of the proposed method is verified through simulations. The simulation results indicate that, in the IEEE-14 node system case study, the accuracy for normal data reaches 97.58%, the accuracy for load step change detection reaches 99.52%, the accuracy for bad data detection reaches 99.92%, and the accuracy under fault conditions reaches 100%.

In LCC-HVDC sending-end AC systems, cascading failures combined with the dynamic response of wind turbines (WTs) can lead to HVDC commutation failures. The resulting transient voltage disturbances cause WT tripping in sending-end systems. Cascading failures that involve the interaction between WTs and HVDC significantly limit the wind power transmitted by HVDC systems. This paper proposes an online cascading failure screening method based on gradient boosting decision tree (GBDT) for HVDC sending-end systems with large-scale WTs. First, a confidence level-based WT tripping model is proposed for cascading failure risk assessment considering a typical cascading failure propagation pattern. Then, Monte Carlo tree search is improved using a contrastive pruning technique to generate evenly distributed samples of cascading failures offline. Dynamic insecure scenarios are quickly identified using an improved support vector machine. Finally, GBDT is utilized to screen for cascading failures online by predicting subsequent high-risk failures using operating features. A dynamic weighting technique is proposed for GBDT to improve the fault prediction accuracy. Simulation results of a modified New England test system and the Ningxia provincial power grid in western China demonstrate that the proposed method can quickly screen for cascading failures considering the dynamic interaction between WTs and HVDC.

Coordinating various controllable distributed resources to reduce network losses is crucial to the secure and economical operation of modern power systems. This paper proposes a bi-level optimization model for power system loss reduction based on “source-grid-load-storage” coordinated optimization. The upper level aims to minimize the total annual planning cost of the system, determining the location and capacity of distributed photovoltaic systems, energy storage devices, and electric vehicle charging stations. The lower level aims to minimize the load curve smoothness and node voltage deviation of the distribution network, optimizing intraday operation strategies. For this complex optimization problem, this paper designs a particle swarm optimization (PSO) algorithm with adaptive weights and improved evolutionary strategies. The simulation results of case studies demonstrate that the proposed method has significant loss reduction effects in distribution networks of various scales and complexities. The algorithm performance comparison results show that the improved particle swarm algorithm outperforms traditional algorithms in terms of solution quality and computational efficiency, providing an effective solution for the coordinated optimization of “source-grid-load-storage”.

Under high distributed generation (DG) penetration, state estimation of active distribution networks (ADNs) faces three core bottlenecks: measurement sparsity, system nonlinear coupling, and non-Gaussian noise interference, severely compromising grid operational situational awareness accuracy and reliability. To address these issues, this paper proposes a forecasting-aided state estimation (FASE) framework integrating Group-Rational enhanced Kolmogorov-Arnold Network (GLKAN) and Generalized Robust PCA Crossformer (GRformer) to synergistically optimize high-fidelity pseudo-measurement generation and accurate state estimation. First, to resolve the parameter redundancy and insufficient dynamic representation of traditional multi-layer perceptrons (MLPs), the GLKAN is proposed to reconstruct the feature extraction unit. Leveraging GLKAN's function decomposition capability and group-rational activation-based adaptability, this approach enables efficient multi-dimensional nonlinear dynamic characterization of ADNs while achieving effective parameter lightweighting. Second, to fundamentally suppress non-Gaussian noise interference on spatiotemporal feature modeling, the GRformer module is built. Its end-to-end low-rank valid component-noise residual separation significantly enhances pseudo-measurement anti-interference. Finally, GLKAN-GRformer is integrated with adaptive interpolation-based strong tracking extended Kalman filter (AISTEKF), which forms a full-link state estimation scheme covering feature extraction, noise suppression and state correction. Experiments on IEEE 33/118-bus distribution systems demonstrate that the proposed method outperforms traditional methods in estimation accuracy and non-Gaussian noise robustness, proving its technical superiority and engineering value in complex ADNs.