Abstract:With the rapid development of emerging technologies such as the Internet of Things and artificial intelligence, the threat of network security is becoming more and more complex, and the network security exposure of The power industry is expanding rapidly. In order to prevent and control network security efficiently and accurately, this paper proposes a new management mode based on RPA, machine learning and other digital technologies, which cooperates with the prevention and control of " people, things, environment and management ". This mode optimizes the network security monitoring and emergency response mechanism, realizes the efficient management of human-machine collaboration, realizes the prevention and control extension to simulate manual operation with RPA technology, and automatically performs command monitoring, sensitive information investigation and early warning reminders. The prevention and control kernel simplifies the security baseline reinforcement process and adapts to the requirements of the domestic digital ecological policy by using the security reinforcement tool based on the unified operating system and the " one area one measure " emergency disconnection scheme. The baseline reinforcement time is shortened from 10 minutes to 21 seconds, which effectively improves the reinforcement efficiency. The research and application results show that the model realizes the three-level linkage of management, technical personnel and user personnel, the response state of network security instructions all day, and the machine generation of network prevention and control part of the business. It further simplifies the workflow, significantly improves the monitoring efficiency and accuracy, and saves human resources.