Research on real-time reliability evaluation of CPS system based on machine learning

This paper studies the reliability of CPS software system and proposes a real-time evaluation method of CPS software reliability based on machine learning. First, use complex networks to identify key points in the network topology of the CPS software system. Unsupervised learning classification by fast density clustering algorithm to classify the importance of nodes can be effectively applied to the importance evaluation of nodes in CPS software system and support the planning of CPS software system Secondly, a real-time CPS reliability automatic online evaluation method is proposed. This method uses machine learning ideas to build an evaluation framework, design an online queuing algorithm, and implement real-time online analysis and evaluation of CPS reliability. Preventive measures ensure that the system operates normally and without interruption, which greatly improves system reliability. Finally, simulation results verify the effectiveness of the evaluation method and its broad application prospects.