Resilience-Aware Energy Management of Microgrids Incorporating Dynamic Thermal Line Rating

The growing infiltration of renewable energy sources, these bidirectional flows, and disturbance-induced operation have posed a considerable challenge in the distribution-level microgrids, where the constraints of the typical energy management system (EMS) based on static network constraints have been exposed. Specifically, the thermal characteristics of distribution lines are so frequently overlooked. The paper suggests a resilience-conscious microgrid energy management system that directly includes dynamic thermal line rating (TLR) as one of the key operational constraints. A hybrid learning-optimization approach is designed, with the integration of Long Short-Term Memory (LSTM) based forecasting of renewable generation, load demand, and conductor temperature with a mixed-integer linear program (MILP) based EMS. The suggested formulation combines time-varying thermal limits that are associated with normal, emergency, and recovery operating conditions and allows consistent conditions. Extensive simulation experiments are undertaken on a distribution level of the microgrid during grid outages, renewable overload, generator failure, and post fault recovery conditions. The proposed solution removes thermal violations completely, serves more total load during disturbances by up to 9% and sheds non-critical loads fewer times by about 40% than conventional EMS formulations. In addition, the proposed scheme reduces the realistic total cost of operation by 21% of the proposed EMS over the approach of applying the static limits to the problem. The results determine thermal line rating as a key enabler of resilient microgrid energy management.