Fei Feng, Lizhi Wang, Ziqian Liu · May 20, 2026
A formal verification method has been developed to assess the shock responses of shipboard microgrids' neural controllers, providing a set-based DAE model compatible with set propagation and a DAE-embedded bound propagation approach. The method computes tight envelopes of all possible neural control output, allowing for effective certification of SMG control performance under uncertain disturbances. This verifies the safety and reliability of neural control in handling nonlinear dynamics in SMGs.
Why This Matters
This paper is relevant to power system engineers as it addresses the control and verification of shipboard microgrids, which can be applied to other grid-scale applications such as remote or islanded microgrids, and provides insights into managing transient shocks, a critical consideration for grid operators and planners facing increasing levels of uncertainty. The methods developed could inform the design and operation of more resilient and adaptive power systems.
Lizhi Wang, Fei Feng, Ella Chou et al. · May 20, 2026
A passivity-based control framework is presented for AC-DC converters in DC data centers to ensure stability under grid and load disturbances. The proposed method uses a Port-Hamiltonian formulation to shape stored energy and inject virtual damping, making the converter behave like a passive system. Simulation studies demonstrate that the controller achieves smaller voltage deviations, faster recovery, and superior robustness.
Why This Matters
This paper matters for power industry professionals as it presents a resilient control framework for DC data centers, which can enhance grid resilience and stability under disturbances such as grid faults or load imbalances, ultimately contributing to the efficient integration of non-passive loads into the power system.
Devangi, Ankit Singhal, Yashasvi Bansal · May 20, 2026
A learning-augmented P2P energy trading framework predicts DSO responses to proposed trades, enabling prosumers to self-assess and refine their decisions without sharing their trades. This allows for reduced transaction overhead, alleviated DSO burden, and preserved information privacy. The proposed model is validated on a modified IEEE 33 bus distribution power system with interconnected microgrids.
Why This Matters
This paper matters for power industry professionals as it proposes a novel framework for predicting the distribution system operator's response to peer-to-peer energy trades, enabling prosumers to make informed decisions and optimize market efficiency. This can be particularly useful in utility planning and capacity markets, where accurate forecasting of DSO responses is critical to ensuring grid resilience and stability.
Electrek