A virtual simulation platform for resilience enhancement in new distribution systems

In recent years, extreme disaster events have exhibited a trend of higher frequency, wider impact, and stronger intensity. This has imposed stringent requirements on the secure and reliable operation of distribution systems. Concurrently, the large-scale integration of distributed generation (DG) enhances operational flexibility but substantially complicates post-fault behaviors. The most notable of these post-fault behaviors involve the distribution of fault currents and the consequent decisions on fault isolation and service restoration, particularly when multiple concurrent faults and dynamically changing network topologies are present. Despite the prevalence of the term “distribution system resilience” in academic discourse, which characterizes the ability to prevent, withstand, respond to, and recover from extreme disturbances, the existing curriculum in electrical engineering remains predominantly focused on conventional subjects such as power flow, short-circuit calculation, and relay protection. There is a conspicuous absence of an integrated training environment that would link resilience theory to engineering actions. Therefore, the objective of this study is to develop a virtual simulation experimental platform that enables students to comprehend the full-chain and multi-stage process of resilience and to practice designing strategies for resilience enhancement in a scenario-driven and operable manner.

The proposed platform is constructed by integrating full-chain resilience concepts (i.e., prevention, mitigation, response, and recovery) with distribution system operational models. It also involves the organization of learning tasks according to the staged evolution of system performance under extreme events. To comprehensively address the resilience workflow, four progressive experimental modules have been implemented: (1) A pre-event preventive proactive islanding simulation is employed, wherein network reconfiguration using remotely controlled switches and manual switches is applied to pre-partition the system into supply islands, with the objective of protecting critical loads from external faults. (2) A fault-current distribution simulation is conducted based on a virtual network-flow representation, which unifies the description of fault-current existence and propagation paths under multi-source, multi-fault conditions and changing switch states. (3) A fault propagation and faulted-section identification module is developed, which explicitly models how automatic tripping of protective switching devices changes topological connectivity and hence delineates the fault influence region. (4) A multi-stage service restoration module formulates post-event isolation and restoration as coordinated topology reconfiguration and supply reachability with multiple switch types, while enforcing that restoration actions must not expand the fault-affected area. The platform facilitates interactive simulation with customizable networks and fault scenarios, providing visualized outputs to ensure the traceability and interpretability of the resilience response process for learners.

The developed platform operationalizes the abstract resilience concept into a coherent sequence of experiments spanning preparation, degradation analysis, and staged recovery. This enables systematic “process-level” learning rather than isolated topic exercises. Module 1 utilizes comparative topology settings to demonstrate how proactive islanding prior to an event can substantially compress the affected region and enhance the continuity of critical-load supply without altering the fault location. Module 2 demonstrates the efficacy of virtual network-flow modeling in representing fault-current distribution at the topological level. It expedites the generation of lucid results concerning propagation paths and associated influence regions in complex DG/multi-fault scenarios. This approach mitigates the modeling and reasoning burden when compared with exhaustive path enumeration, thereby directly supporting subsequent localization, isolation, and restoration reasoning. Module 3 employs a visual approach to illustrate the causal chain from protective-device tripping logic to the emergence of clearly delineated fault impact regions. This module serves to reinforce students’ comprehension of how switching operations influence fault transfer and faulted-section localization. Module 4 delineates multi-stage restoration as an executable sequence of switching operations and resource coordination. The sequence initiates with the “compression” and isolation of the fault region, followed by a stepwise expansion of the restored supply. This approach enables learners to observe the impact of diverse action sequences on restoration scale and tempo.

The integration of full-chain resilience theory with operable distribution system models, embedded within four progressively linked simulation modules, establishes a nexus between theoretical instruction and engineering practice for resilience-oriented education under extreme events and high DG penetration.