微电网
模式(计算机接口)
计算机科学
断层(地质)
估计
控制理论(社会学)
人工智能
控制(管理)
工程类
操作系统
地质学
地震学
系统工程
作者
Peng Zhang,Peng Wu,Jianwei Ren
出处
期刊:Physica Scripta
[IOP Publishing]
日期:2025-08-27
卷期号:100 (9): 095208-095208
标识
DOI:10.1088/1402-4896/adffc7
摘要
Abstract Information leakage and voltage balance issues among subsystems introduce additional uncertainties to the stable operation of distributed DC microgrids. Although effective algorithms for voltage balance exist, comprehensive solutions addressing the aforementioned disturbances are scarce due to privacy protection constraints. To meet these challenges, this paper proposes a privacy-preserving sliding mode control (SMC) method for distributed DC microgrids. First, a mathematical model of a DC microgrid with an energy storage source is established and transformed into a mathematical multi-agent system (MAS) model. Additionally, an output mask privacy-preserving method is employed by introducing an output mask function to obscure system state information while ensuring the protected information converges accurately to the reference value. Subsequently, to enhance the convergence performance of the masked system, a SMC approach based on variable structure is employed. Considering the nonlinearity introduced by the mask function, the traditional SMC is improved and extended to accommodate multi-agent masked systems, enabling the design of more effective sliding mode controllers. Moreover, actuator faults leading to measurement errors are accounted for in the design of the sliding mode controller, enhancing the fault-tolerant capability of the system. Simulation results demonstrate that the proposed scheme not only ensures privacy protection but achieves fast convergence of the sliding surface and control input within average times of 0.38s and 0.65s, respectively, outperforming conventional SMC. Furthermore, the scheme yields smaller steady-state errors, highlighting its superior dynamic and steady-state performance. These simulation results validate the feasibility of the proposed approach.
科研通智能强力驱动
Strongly Powered by AbleSci AI