Digital Substation Security Monitoring Network Intrusion Detection Based on Deep Learning
Xin Wu, Gang Qu, Haochun Jin, Liang Zhang · 2023
In order to improve the security of substation monitoring network, a digital substation security monitoring network intrusion detection method based on deep learning is proposed. Firstly, the architecture of digital substation safety monitoring network is analyzed. Secondly, the collected substation safety monitoring network data are completed, redundant feature reduction and normalization pre-processing. Finally, by using the deep belief network model in deep learning and through the pre-training process, the parameters in the constrained Boltzmann machine can be initialized, the deep belief network can be optimized, the optimal weight and bias can be obtained, and the intrusion classification detection of security monitoring network can be performed. Experimental results show that compared with the existing intrusion detection methods, the detection accuracy of the proposed method is always above 95%, and the lowest false positive rate is only 0.81%.