A Distributed Secure and Robust Power Management Framework for More Electric Aircraft

More electric aircraft (MEA) needs to efficiently manage power flow and maintain a high-quality provision for loads in its electrical power system (EPS). This paper proposes a distributed power management framework for MEA to realize real-time power distribution under device failures and adapt to significantly varying uncertain loads. The framework consists of two parts: fault isolation and reconfiguration, and rolling scheduling. In the first part, EPS topology is represented as an undirected graph. A fault isolation set is formulated to suppress future device failures. After suppressing faults, the EPS network is divided into multiple islanded microgrids with optimal power supply paths, which provides an architectural basis for distributed modeling and scheduling of MEA power management problem. In the second part, each microgrid proposes its multi-stage distributionally robust model, where load uncertainty is considered and fully respects the non-anticipativity in the decision-making process. Based on the introduced piecewise linear decision rule, decision variables can adjust with the observation of load outcomes and have a tighter approximation. The formulated problems are cast into tractable linear programming ones, which are finally solved and scheduled in parallel. The results indicate that the solution speed is accelerated, and the communication costs are reduced.