An IoT-integrated stochastic cloud theory framework for carbon-restricted smart building energy management within the context of next-generation energy performance certificates

Abstract

The transition toward Next-Generation Energy Performance Certificates (EPCs) has created new opportunities for integrating digitalization and intelligent control in smart buildings. In this context, the rapid development of smart building technologies and the Internet of Things (IoT) has intensified the demand for advanced energy management frameworks capable of balancing energy efficiency, economic performance, and environmental sustainability. This study proposes a stochastic carbon-restricted energy management framework for IoT-integrated smart buildings that combine renewable energy sources (RES), combined heat and power (CHP) units, auxiliary boilers, electric vehicles (EVs), and hybrid electrical–thermal energy storage systems. The model aims to optimize the trade-off among cost, carbon reduction, and occupant comfort under emission-constrained operation. To accurately capture the uncertainties in renewable generation, load demand, and electricity price, a Cloud Theory-based uncertainty modeling approach is adopted, enabling the simultaneous representation of randomness and fuzziness in system parameters. The optimization process employs the Boxelder Bug Search Optimization (BBSO) algorithm, which offers strong global exploration and fast convergence. IoT-based monitoring and data interoperability enhance real-time coordination among distributed resources, contributing to the digital transformation envisioned in EPC frameworks. Three operational scenarios are investigated: deterministic operation without carbon limits, deterministic operation under carbon limits, and stochastic operation under Cloud Theory-based uncertainty. All simulations are conducted over a representative two-day (48-hour) winter period with an hourly time resolution to ensure consistent temporal modeling across all scenarios. Results indicate that carbon restriction increases total operating cost by 17.3%, while stochastic modeling adds an additional 4.6% due to risk buffering. Meanwhile, key EPC-relevant indicators show that carbon intensity and primary energy consumption increase moderately, while the self-sufficiency ratio decreases from 61.93% to 55.95%, revealing a clear cost–resilience trade-off beyond static EPC metrics. Nevertheless, the proposed strategy reduces grid dependency by 5.8%, improves renewable utilization by 6.4%, and ensures full occupant comfort across all cases. Comparative benchmarking confirms the superiority of the proposed BBSO over other established optimization techniques, achieving up to a 5.14% reduction in operational cost while maintaining high solution robustness. These findings demonstrate that the IoT-integrated and Cloud Theory-based optimization approach provides a robust pathway toward data-driven, low-carbon, and EPC-aligned smart building management systems.