Hybrid fuzzy-PID control of a nuclear Cyber-Physical System working under varying environmental conditions

Abstract : In real-time control of Cyber-Physical Systems (CPSs), physical process variables are monitored and processed by intelligent controllers for keeping the values of safety parameters between given thresholds. Environmental conditions can affect the system dynamics and also the controller function. This paper presents a hybrid fuzzy-PID (Proportional-Integral-Derivative) controller, which learns the optimal PID controller parameters and adapts them according to the environmental parameters and process variables values. The proposed intelligent controller is applied to respond to the changes and transients in the environmental inlet air temperature of the secondary loop of a Lead-Bismuth Eutectic eXperimental Accelerator Driven System (LBE-XADS), for keeping the secondary average coolant temperature in a range of values between two safety thresholds.