Aiming at the problem of network intrusion feature optimization, a binary particle swarm optimization algorithm based on manta rays foraging algorithm is proposed for feature selection of network intrusion detection, which improves the network intrusion detection rate. Firstly, the network intrusion detection rate is used as the objective function of feature selection, and the network state feature is used as the constraint condition to establish the corresponding mathematical model. Then, the binary particle swarm optimization algorithm of manta rays foraging learning strategy is used to find the optimal feature subset. Finally, the machine learning classifier is used to establish the intrusion detection model. The experimental results show that the model can efficiently query the optimal feature subset, and the intrusion detection rate and efficiency are better than the classical intrusion detection model.