Network Intrusion Detection Method for Smart Grid Based on PCA-ISBO-GRU-AM
Zhiying Wang, Feifei Zhang, Hao Wang, Xiangcong Zhang, Weizhi Lu, Chan Zhang, Lei Wang, Bingjie Wang · 2024
Network communication technology plays an extremely important role in the construction of smart grid, but this also makes the smart grid network security protection has become the key to the operation of the system. Advanced metering infrastructure (AMI) as the core component of the smart grid, the characteristics of its two-way communication makes itself has been facing the threat of network intrusion, and the current detection methods are difficult to effectively detect network intrusion. The current detection methods are difficult to effectively detect network intrusion. In this paper, we propose a deep learning-based network intrusion detection method for smart grid AMI in response to the lack of accuracy of network intrusion detection. Firstly, the AMI system of smart grid is designed, and principal component analysis (PCA) is used to complete the data dimensionality reduction. Then the gate recurrent unit (GRU), which introduces the attention mechanism (AM), is used as the base detection model, and the satin bowerbird optimization (SBO) algorithm, which is improved by the introduction of the risk avoidance principle, is used to realize the hyper-parameter optimization of the GRU, so as to get the network intrusion detection model based on ISBO-GRU-AM. model. The resulting model is finally tested on KDDCup 99 and NSL-KDD datasets to check the detection performance of the model. The experimental results show that the method proposed in this paper can accurately detect the network intrusion of smart grid AMI and distinguish the type of intrusion, which has obvious advantages over other methods, and is conducive to promoting the progress of AMI network security protection technology and ensuring the network security of smart grid.