基于云服务的虚拟电厂优化调度与隐私保护

Privacy-preserving Optimal Dispatch of Virtual Power Plant Based on Cloud Service

虚拟电厂通过先进的通信、计量、控制等技术聚合不同的灵活性资源,实现资源综合协调优化控制,是提高电网安全水平、降低用户用能成本、促进新能源消纳的重要措施。而基于云服务的计算框架将虚拟电厂优化调度的求解转移至云端服务器完成,在降低本地计算压力和计算资源投资成本的同时也带来了新的隐私安全问题。为了应对云计算过程中云端服务器与园区之间信息交互存在的隐私泄漏风险,文章研究并提出了面向云服务的虚拟电厂优化调度过程中基线计算环节的信息伪装机制,利用基于线性映射的信息伪装算法将原优化问题的参数及变量信息进行伪装,在有效保证云端模型求解正确性的同时,可靠地保护园区数据隐私。通过对算例优化结果以及信息伪装特性的分析验证了所提模型和方法的有效性。

Virtual power plant aggregates different flexible resources through advanced communication,metering,control and other technologies to achieve comprehensive coordination and optimal control.It is an important measure to improve the security level of the power grid,reduce energy consumption cost,and promote the consumption of new energy.The framework based on cloud services transfers the calculation of optimal scheduling of virtual power plants to the cloud server,which not only reduces the local computing pressure and computing resource investment cost,but also brings new privacy and security issues.To deal with the risk of privacy leakage in the information interaction between cloud servers and virtual power plants in the process of cloud computing,this paper studies and proposes an information masking mechanism for optimal scheduling of virtual power plants oriented to cloud services.The parameters and variable information of the problem are obscured,which can effectively ensure the correctness of the optimal solution and reliably protect the data privacy of the virtual power plant.The effectiveness of the proposed model and method is verified on a practical case study through the analysis on the outcome of optimization results and the features in information masking.