A Resilient and Privacy-Preserving Cloud-Edge Distributed Method for Reliable Economic Dispatch in Cyber-Physical Energy System

With the rapid development of 5G and IoT technologies, privacy concerns caused by passive eavesdropping attacks have become increasingly prominent. Firstly, an economic dispatch model for information-energy systems considering the Cloud-Edge-Devices architecture is established, with refined modeling of energy transmission characteristics and losses. A KL-divergence-based differential privacy mechanism is designed for the data acquisition stage. By introducing Gaussian noise with standard deviation correlated to the marginal cost, the privacy of terminal devices during data collection and transmission is effectively preserved. Secondly, a privacy-preserving distributed consensus economic dispatch algorithm (PP-DCED) is proposed. Local secret functions are embedded into the multi-energy management controllers at the edge layer. Combined with a disturbance gain feedback mechanism and a Gaussian decaying noise-based iterative algorithm, the proposed method ensures privacy protection during data exchange among edge devices. Finally, a hierarchical privacy protection scheme is achieved across the Cloud-Edge-Devices architecture. Hardware-in-the-loop simulation results demonstrate that compared with traditional centralized methods, the proposed approach reduces the privacy disclosure risk by 32.5% and improves communication efficiency by 24.8% while maintaining a high convergence accuracy of 0.001. This confirms that the proposed method can effectively balance the trade-off between privacy requirements and economic dispatch performance in complex energy ecosystems.