In this research paper, an adaptive and intelligent protection scheme is developed that brings selectivity and sensitivity to the conventional overcurrent relays considering the changes in grid topologies, changes in grid operation modes, and changes in short-circuit behavior due to the contributions from distributed energy resources in the medium-voltage distribution grids. A quasi-static medium-voltage benchmark grid model is used in the Python-based power system tool pandapower. Definite-time overcurrent relay devices are extended with a communication signal-based tripping scheme suitable for a decentralized communication architecture. Current phasor angles are compared between two primary relays and the result is fed into a tripping logic to achieve selectivity of the primary protection. Furthermore, communication between the primary and backup relays is established, extending the communication signal-based tripping scheme. It analyzes in which cases the proposed communication-based protection scheme’s ability to isolate the faulty section after short-circuit events is comparable to conventional schemes based on distance relays that require additional voltage measurements or differential relays that require dedicated high-speed communication. Therefore, the pre-existing and installed communication capabilities can be used to avoid the need for additional measurement hardware or communication technology.

Multiple-fault events can initiate overload propagation and cascading outages, resulting in severe load loss and reduced system resilience. Therefore, beyond conventional protection concepts based on the (n − 1) criterion, there is also a need to address multiple-fault events to minimize loss of load. This paper presents an optimized overload tripping scheme to mitigate cascading outages in high-voltage grids under multiple-fault conditions, where selected line switches or circuit breakers are opened in a controlled manner to isolate limited grid sections, minimize interrupted load, and prevent further overload propagation. The method combines inverse definite minimum time relay modeling with a heuristic graph-search algorithm implemented in pandapower to identify feasible switching actions that minimize load loss while preventing overload propagation. The approach is demonstrated on SimBench high-voltage urban and mixed benchmark grids under double-line fault scenarios. In the urban grid, the proposed scheme reduces the maximum load loss from 34.0% to 2.4%, while in the mixed grid, the reduction is from 50.3% to 5.2%. A SAIFI-inspired resilience proxy is introduced to quantify the reduction in customer/load interruptions, showing a resilience improvement factor of about 3.6 for cascading scenarios. In addition, thermal inertia analysis indicates that corrective switching must be completed within approximately 5 min to remain within line-temperature limits. The analysis is based on quasi-steady-state power-flow and relay simulations; transient stability effects are outside the scope of this study. The results demonstrate that the optimized overload tripping scheme is a promising adaptive protection strategy for improving grid resilience under severe contingency conditions.