Consumer-Electronics-Based IoT Data Fusion for Dynamic Operating-Region Correction in Distributed Electricity-Heat Coupled Systems

High renewable penetration in distributed electricity–heat coupled systems (DEHCSs) introduces model mismatch and boundary uncertainty under equipment aging, parameter drift, and nonstationary fluctuations. This paper develops a consumer electronics–enabled IoT information-fusion intelligent system for dynamic operating-region correction. On the perception side, consumer electronics devices complement grid-grade sensors to provide dense source–network–load observations and enhance the observability of thermal-demand dynamics and user behavior. On the decision side, the proposed intelligent system integrates heterogeneous data streams and constructs a joint nominal operating region using Minkowski-sum-based set operations under electro-thermal coupling and regulation constraints. Subsequently, a robust symplectic geometric mode decomposition (R-SGMD) method maps nonstationary time-series data of wind/PV outputs and thermal demands into a structure-preserving phase space, where high-frequency modal energy deviations are extracted to quantify real-time uncertainty. These features drive online tightening/relaxation of renewable availability bounds and thermal-demand uncertainty envelopes, and are embedded into a bilevel optimal dispatch model to balance security margins and economic performance. Case studies show that the proposed intelligent system improves wind/PV integration while avoiding security blind spots caused by static operating boundaries.