Vulnerability analysis of demand-response with renewable energy integration in smart grids to cyber attacks and online detection methods

The two-way information exchange between customers and the utility in smart grids enables demand-response programs of customers and the integration of distributed renewable energy resources. However, this also makes the demand-response programs vulnerable to cyber attacks. In this paper, we study cyber attacks that target customers' demand-response programs in smart grids by injecting false consumption and generation information. Then, as a countermeasure, an online detector based on convolutional neural networks is designed to detect the cyber attacks and mitigate impacts. The vulnerability of power distribution systems with and without the proposed detector is analyzed with reference to a case study concerning the IEEE 34 bus test feeder. The results show that the power distribution systems is vulnerable to the studied cyber attack and the proposed detector can achieve high accuracy and mitigate the impact of cyber attacks with fixed change rates, whereas the attacks with variable change rates are inherently challenging to detect.