计算机科学
控制理论(社会学)
控制(管理)
控制工程
自适应控制
控制系统
工程类
弹性(材料科学)
稳健性(进化)
弹道
作者
Qianhui Ma,Pengcheng Hu,Abhisek Ukil,Akshya Kumar Swain
出处
期刊:Applied Energy
[Elsevier BV]
日期:2026-08-20
卷期号:426: 128691-128691
标识
DOI:10.1016/j.apenergy.2026.128691
摘要
In renewable-integrated multi-microgrid (MMG) systems, constituent microgrids differ in state-of-charge (SOC), rated capacity, and load margin. Existing consensus-based coordination either ignores this heterogeneity or weights only SOC within single-storage configurations, leaving the joint role of capacity and load margin unaddressed. This paper proposes a multi-dimensional weighted adaptive consensus (MD-WAC) control strategy for MMG systems equipped with a hierarchical hybrid energy storage system (HESS) of batteries and supercapacitors. A sigmoid weighting parameterized by SOC, capacity, and load-margin deviations modulates the inter-microgrid Laplacian, and a frequency-decoupled HESS layer assigns low-frequency power to batteries and high-frequency transients to supercapacitors. Closed-loop asymptotic stability follows from singular perturbation analysis on the three-timescale separation among supercapacitor, consensus, and SOC dynamics. The strategy is validated on a three-node hardware testbed with photovoltaic panels, lead-acid batteries, and supercapacitor modules, and on a 50-node numerical study. Under load step changes, renewable fluctuations, and topology switching, MD-WAC keeps the DC bus voltage deviation within ± 1 % , achieves a 0.22 s battery current response, redistributes power according to each microgrid’s SOC, capacity, and load margin, and retains a 51.6% low-frequency share of battery energy against 2.5%–32.4% for four baseline methods. The per-iteration computation grows by less than 6 % from N = 3 to N = 50 , against a 303 × increase for centralized MPC.
科研通智能强力驱动
Strongly Powered by AbleSci AI