Performance and reliability improvement of cyber-physical systems subject to degraded communication networks through robust optimization

The study of cyber-physical systems (CPSs) is a rapidly developing field because of their broad application in different areas. However, the degradation, i.e., time delay and packet dropout, that originated from the cyber layer-communication networks would deteriorate the performance of the physical layer. Hence, the internal dependence and reliability of such a complex and safety-critical systems must be analyzed. In this paper, we describe a CPS by using time-varying models of its main components and realize it in TrueTime simulator. The Monte Carlo simulation is applied to estimate the system reliability. The parameters of the discrete-time proportional-integral-derivative (PID) controller are tuned to improve the performance and reliability of the CPS. Considering the controller parameters that undergo perturbation when in hardware implementation, we use particle swarm optimization to search for a robust PID controller that can ensure a satisfied control performance against parameter perturbation. An industrial heat exchanger is studied to illustrate the correctness and efficiency of the proposed method.