Smart Meter Intrusion Detection Based on APSO-DBN Model

With the continuous development of network technology and information technology, more and more attention has been paid to the network security of smart meters. When facing the network intrusion data of smart meters, the data has large-scale, high-dimensional and non-linear intrusion detection problems. This paper proposes a model about the network intrusion model, which is based on Adaptive Particle Optimization (APSO) Deep Belief Network (DBN). First, construct a classification model based on DBN algorithm. Next, accelerate the optimization of DBN by using APSO algorithm, and calculate the number of hidden layer nodes to obtain the DBN model structure of the best hidden layer. Finally, construct the intrusion detection model of smart meter based on APSO-DBN.