CPS rare event probability prediction method based on statistical model inspection

Abstract

The invention discloses a CPS rare event probability prediction method based on statistical model inspection. The system is based on a statistical model inspection method and is composed of a samplingmonitor, a protocol verifier and a probability predictor. The method comprises the specific implementation steps that a sampling monitor adopts an importance random sampling algorithm to randomly sample and generate a CPS system state feature sequence described based on a hybrid automaton; a protocol verifier based on linear temporal logic can generate a system visible characteristic state sequence and a verification result corresponding to protocol satisfaction, and the verifier can take the verification result as a training sample set of a probability predictor based on an HMM implicit Markov model; on the basis of sufficient sample set learning, the probability predictor calculates the probability that the subsequent state of the system is a rare event state, and outputs the result, inaddition, the method also has the function of correcting the prediction capability of the predictor by using the probability prediction result.

Description

technical field [0001] The invention belongs to the field of cyber-physical system control, and in particular relates to a method for predicting the probability of CPS rare events based on statistical model testing Background technique [0002] With the gradual development of "Industry 4.0" and "Made in China 2025", the supply, manufacturing, and sales processes in production are all digitized and informatized through cyber-physical systems to achieve rapid and effective applications. The Cyber-Physical System (Cyber-Physical System) realizes the closed-loop path between the physical space and the information space. In essence, it is to transmit various data in the physical space to the information space, and use the powerful analysis and processing capabilities of the information space to find the It is the process of realizing the best decision to optimize the allocation of physical space resources and applying the decision to the physical space in the form of data. As a ...

AI Technical Summary

Problems solved by technology

[0005] CPS is an open system, and uncertainty is an inherent characteristic of this type of open system. The uncertainty of CPS may lead to system security problems. Once a security problem occurs, it will have catastrophic consequences. Serious Threat to people's lives and property safety

Therefore, in CPS verification, it is a huge challenge to be able to verify the impact of uncertain behavior on system security.

Images