With the digital transformation of industrial enterprises, the energy industry environment faces increasingly complex network security and information security issues. Based on Kubernetes cluster management system, the article establishes an emergency security response system for the energy industry and designs a series of functional modules for data collection, security detection and emergency response. In order to improve the network intrusion detection capability of the emergency security response system, this paper proposes a network intrusion detection model based on 1DCNN-BiGRU. The 1DCNN and BiGRU models are used to obtain the spatial and temporal features of multimodal data, respectively, and the fusion of the spatial and temporal features of multimodal data is realized through the full connectivity layer, which then realizes the network intrusion detection of the emergency security response system. The study shows that the weighted average F1 score of the 1DCNN-BiGRU model in network intrusion detection is up to 0.9335, and the acceleration ratio of the emergency response system in the energy industry is up to 143.91, which can effectively realize the system dynamic load balancing. Fully exploiting the changing characteristics of multimodal data and integrating deep learning to promote the energy industry emergency safety response capability.
