As the process of digital transformation accelerates, cybersecurity defense is facing increasingly severe challenges. To enhance their defensive capabilities, some cybersecurity companies have adopted a collaborative monitoring strategy using multi-source situational awareness devices. However, in practical applications, they have encountered prominent issues such as a massive volume of alert data and high false alarm rates. In response to this problem, this study proposes an innovative and optimized algorithm framework. This algorithm integrates core components including a sliding window mechanism, a correlation alignment algorithm, and the Dempster-Shafer fusion algorithm. The main innovations of this algorithm lie in bypassing the data annotation process entirely and directly applying statistical learning methods to extract features related to alert credibility and attention-worthiness, as well as the establishment of a dual-matrix-driven mechanism to achieve intelligent fusion of alert data. Through these algorithmic innovations, it achieves a reduction in the number of alerts and an improvement in accuracy and attention-worthiness. To verify the algorithm's effectiveness, researchers selected appropriate subjects for experimentation. The experimental results demonstrate that the algorithm can significantly compress the number of alerts, substantially enhance detection accuracy and attention-worthiness, effectively integrate the advantages of multi-device monitoring, and expand the coverage of security defense.

Superconducting electrodynamic suspension (SC-EDS) train, characterized by its large levitation gap, high lift-to-drag ratio, and self-stabilization features, has significant application potential in high-speed and ultra-high-speed maglev transportation. However, the modeling for the SC-EDS train poses significant challenges in balancing accuracy with computational efficiency. This paper proposes a physics-informed neural network (PINN)-based modeling method for the SC-EDS train. Focusing on the dynamic circuit equations governing the levitation/guidance (L/G) coils and onboard SC coils, the method leverages PINN to rapidly solve for the currents in the L/G coils, thereby enabling real-time computation of electromagnetic forces critical for dynamic modeling of the train. By embedding the dynamic circuit equations into neural networks as physical constraints, the model achieves high-fidelity approximations of the currents while drastically reducing computational costs. This approach not only enhances the feasibility of real-time dynamic modeling but also lays a foundation for stability analysis, system optimization, and advanced control strategies in SC-EDS applications.

User-side energy storage systems (USES) play a crucial role in smart grids, renewable energy integration optimization, and demand-side management (DSM). However, due to the complex dynamic interactions among battery management systems (BMS), energy management systems (EMS), and power conversion systems (PCS), traditional simulation methods struggle to comprehensively and accurately capture the multi-scale operational characteristics of USES. To address this issue, a hybrid simulation framework is proposed in this paper, integrating equivalent circuit modeling (ECM), agent-based modeling (ABM), and discrete event simulation (DES) to improve simulation performance. Simulation results demonstrate that the proposed framework could effectively model electrochemical battery characteristics, inter-device interactions, and event-driven system dynamics, providing robust support for scheduling strategy optimization and professional training in energy storage operations.

Clustering is a popular method to deal with the problem for mode identification of multimode processes. Unlike traditional distance-based clustering methods, in this paper, a new correlation-based bi-partition hierarchical clustering (CBHC) method is proposed, which classifies the observations according to their correlation relationships rather than their distances. Motivated by an existing correlation-based mode identification method, a modified similarity matrix is first given by introducing normalization and sparseness into that of the existing method, then a bi-partition hierarchical clustering is used to further classify the observations. The proposed method can remove two strict assumptions required by the existing correlation-based mode identification method, i.e. the orthogonal assumption and the assumption that the number of modes should be known in advance. The merits of the proposed method are proved through two numerical examples.