In view of the fact that current safety risk assessment models for industrial cyber physical systems(ICPS) rarely consider the impact of dynamic process of the system on the accuracy of the assessment, this paper proposes a quantitative evaluation model for the dynamic performance analysis of security risk in the ICPS. Firstly, the Bayesian network is used to model the intrusion process of the attack in the cyber layer, and the probability of the successful intrusion of the network attack is calculated. Then, under the premise of successful attack, the Kalman state observer is used to observe the state of the controlled object in real time, the dynamic performance of the system is studied, the performance loss of the system is quantitatively analyzed, the impact of the attack on the system from the perspective of economic loss is quantified, and the dynamic assessment of system security risk based on the probability of successful attack is realized. Finally, the running state of the boiling water power plant model under attack is simulated using Matlab. The results show that the model can effectively assess the risk of ICPS.