Reliability evaluation of Markov cyber–physical system oriented to cognition of equipment operating status

With the rapid development of computing, communication, and control technologies, cyber–physical systems that integrate physical space, information space, and social space have emerged and are widely used in various important infrastructures. This article introduces the composition and basic algorithm of the hidden Markov model, and gives the mathematical description of the hidden Markov model. Since the hidden Markov model can deduce the hidden state of the observation object through the observed feature values, a device operating state cognition scheme based on the hidden Markov model is proposed. A method for analyzing cascading failures is proposed, and the critical threshold value of cyber–physical system under random attack is obtained. It is verified by simulation experiments, and the changes of system critical thresholds under different network parameters are compared and analyzed. We mainly use several sets of simulation experiments to verify the reliability of the critical threshold, and then verify near the critical threshold. Before simulating the cascading failure process, we first construct two random networks based on the average degree and the number of nodes. According to the previous description of the cyber–physical system model, a node in network B is randomly connected with three nodes in network A, so that the two networks are connected together to form a coupled system. Random attack or failure is represented by randomly deleting nodes. In the simulation experiment, we will simulate the process of cascading failure at each step, and after each step of cascading failure, we output and save the number of remaining nodes. When no nodes in the two networks are deleted, the cascading failure will stop, and then we will verify the critical threshold through the data obtained from the analysis. This provides the support of related theories and methods for the design of stable and reliable cyber–physical systems.