Scheduling Strategy for Autonomous Balancing inDistributed PV Substation Areas under FaultConditions Considering Uncertain BehavioralResponses of Bounded Rational Users

Introduction: Demand Response (DR) has emerged as a critical solution for mitigating voltage violations, line overloads, and cascading failures in distribution networks with high penetration of distributed Photovoltaics (PVs) following fault clearance. However, the inherent bounded rationality, heterogeneity, and uncertainty in user response behaviors, coupled with the stochastic nature of distributed PV outputs, pose significant challenges to achieving reliable post-fault power balance. To address these issues, this paper proposes a distributed autonomous balancing optimization strategy for distributed PV fault scenarios, incorporating both user-side bounded-rational response uncertainty and source-side output uncertainty. Methods: An evolutionary game model on social networks characterizes bounded-rational user responses; a robust optimization framework with polyhedral uncertainty sets models source-load uncertainties; an accelerated distributed algorithm based on Alternating Direction Method of Multipliers (ADMM) meets real-time post-fault rescheduling requirements. Results: The simulated results show that additional incentive costs increase rescheduling expenses, while all node voltages remain within safe operational limits, preventing further faults and ensuring distribution network security, under the proposed bounded-rational demand-response rescheduling strategy. Discussion: The strategy effectively addresses source-load uncertainties and highlights DR’s role in post-fault balance, while its limitation lies in increased costs from additional incentives Conclusion: The proposed strategy can ensure power balance in distribution networks with high penetration of distributed PVs under fault conditions, reduce power consumption costs, and meet the time requirements for load rescheduling after faults.