Research On Risk Assessment Method Of Industrial Robot System Based On Improved Petri Net

The new generation of industrial robots requires not only good and reliable performance of control and servo system,but also better openness,continuou s open interface and increasing network connection,which results in a dramatic increase in the frequency of hackers attacking industrial robot system and the losses caused.As a kind of information security evaluation method,risk assessment technology can analyze the harm to the system qualitatively or quantitatively.At present,the industrial robot system lacks a set of feasible and perfect risk assessment methods,and the risk assessment of other related industries lacks to consider the correlation between risks,which affects the accuracy of risk assessment results.To solve the above problems,this paper presents a risk assessment method for industrial robot system based on improved Petri net on the basis of analysis of the composition and safety of industrial robot system.Firstly,the network architecture and composition of the industrial robot system are analyzed,and the potential attack ways and security requirements of the industrial robot system are determined.Based on the communication message network address discovery,the industrial robot system assets are identified and the important assets in the system are assigned values.The information security threats faced by industrial robot systems are analyzed and assigned.For the vulnerability of industrial robot system,a general vulnerability scoring system is used and assigned.The risk calculation method of isolated equipment in industrial robot system is given.Secondly,in order to analyze the correlation of isolated equipment risk in industrial robot system,the attack chain of industrial robot system is described by Petri net theory,and an attack chain model of industrial robot system based on Petri net is constructed.To reduce the complexity of the attack chain model of industrial robot system,an object-oriented Petri net is used to simplify the attack chain model of industrial robot system,and an approach to modeling the attack chain of industrial robot system based on ordered asset search is presented.In order to analyze the vulnerability and safety state of industrial robot system during attacker attack,a method for predicting the vulnerability and safety state of industrial robot system based on Bayesian network and Petri network is presented.Then,based on the attack chain model of industrial robot system,the relationship between the change of vulnerability security state and the dynamic risk elements is studied,and a calculation method for the safety assurance capability of industrial robot system is presented.According to the composition and safety requirements of industrial robot system,the safety index of industrial robot system is determined,the safety index system of industrial robot system is constructed by principal component clustering,and the safety index is simplified by Re lief F algorithm.The weight of each information security index of industrial robot system is quantified by combined weighting method,and the safety assurance capability of industrial robot system is evaluated according to the gray correlation TOPSIS algorithm.Finally,the connectivity of the attack chain model of the industrial robot system is analyzed,and an attack chain model path search algorithm is presented,which achieves the extraction of attack paths within the attack chain model of the industrial robot system.An attack path risk assessment method for industrial robots based on path impact degree is presented,and the Q-Learning algorithm is used to validate the attack path assessment method.The overall risk assessment method of industrial robot system is given,and the actual industrial robot system is evaluated.