To effectively guarantee a secure and stable operation of a smart substation, it is essential to develop a relay protection system considering the real-time online operation state evaluation and the risk assessment of that substation. In this paper, based on action data, defect data, and network message information of the system protection device (PD), a Markov model-based operation state evaluation method is firstly proposed for each device in the relay protection system (RPS). Then, the risk assessment of RPS in the smart substation is carried out by utilizing the risk transfer network. Finally, to highly verify the usefulness and the effectiveness of the proposed method, a case study of a typical 220 kV substation is provided. It follows from the case study that the developed method can achieve a better improvement for the maintenance plan of the smart substation.

A permanent fault identification scheme based on LCC signal injection for high-voltage direct current (HVDC) systems is proposed to avoid secondary damage when it recloses to a permanent fault. Firstly, using the fault control ability of LCC, the additional control strategy is applied to the trigger angle of LCC to realize signal injection. The frequency, duration, and amplitude of the injection signal are analyzed and determined, and a signal injection strategy based on LCC is proposed. Secondly, the differences in voltage after signal injection under different fault properties are analyzed under the distributed parameter model. There is a significant difference in the amplitude of the measured voltage at the local end and the calculated voltage at the remote end under different fault properties due to differences in line models. Finally, a normalized area differential is constructed based on the above amplitude difference to realize permanent fault identification. PSCAD/EMTDC simulation results show that the proposed scheme utilizes single end data and is not affected by data communication. There is no need to set a threshold through simulation, and it can reliably identify permanent faults under 400 Ω fault resistance and 40 dB noise. It is suitable for line lengths of 1500 km and below.

The fault current in MMC HVDC systems suffers from a quick rising rate and no zero-cross problems, which makes it hard to cut off and has become the main problem when promoting MMC dc systems. Among the different types of faults, the pole-to-pole fault is the most serious one. Hence, to clarify the discharging mechanism during pole-to-pole faults and determine the influential factors, this paper proposes a simplified modeling method for MMC HVDC fault current impact factors analysis. The proposed method uses the simplified MMC circuit when faults happen, and the frequency domain model is then established for fault current analysis. Considering that the fault period concerned always is less than 10 ms, the high-frequency part impedance characteristics of the established model are carefully studied when device parameters are varied. The smoothing reactor has a significant influence on the high-frequency impedance, while the arm inductance and arm resistance have much less impact on high-frequency impedance. This demonstrates, theoretically, that the pole-to-pole fault current is basically influenced by the smoothing reactor rather than other impact factors. The calculation results based on the model also prove that fault current will change a lot if the smoothing reactor’s parameters are different. The analysis verifies the accuracy of the equivalent model and supplies a theoretical analysis tool for MMC HVDC fault current analysis.

With the increasing scale and complexity of power systems, the Security and Stability Control System (SSCS) plays a vital role in ensuring the safe operation of the grid. However, existing SSCS implementations still face many limitations in cross-regional coordination, control precision, and risk prediction. Establishing the digital simulation model is an effective way to verify the control policy of SSCS. This paper proposes a neural heuristic task scheduling method based on deep reinforcement learning (DRL) to schedule the simulation tasks. It models the task dependencies of SSCS as a directed acyclic graph (DAG) and then dynamically optimizes task priorities and resource allocation through deep reinforcement learning. The method introduces multi-head attention and heterogeneous attention mechanisms to effectively capture complex dependencies among tasks, enabling efficient multi-core task scheduling. Simulation results show that the proposed algorithm significantly outperforms traditional scheduling methods in terms of makespan, load balancing, and resource utilization. It can also adapt to dynamic changes under different task scales and multi-core environments, demonstrating strong robustness and scalability.