Multi-agent based fractional order control system for 13C isotope separation column cascade

Abstract

The 13C isotope separation column cascade system is a very rare and complex process which can become even dangerous if it is not operated in a proper manner. The development of an efficient supervisory and control system still represent a challenge due to the number of interconnections and interdependencies between the process variables. The objective of this paper is the development and the implementation of a multi-agent system (MAS) to achieve the efficient control of the 13C isotope separation column cascade. The multi- agent system was implemented using the JADE (Java Agent DEvelopment Framework) platform which allows the parameterization of the characteristics of each agent running on the system. The JADE platform is compliant with the FIPA specifications for the interoperability of heterogeneous agents. The operational model of the isotope separation column cascade is implemented in Matlab/SIMULINK and is linked to the multi-agent control system operating in JADE using MACSimJX (Multiagent Control Using Simulink with Jade Extension). The process output variables from the sensors are sent to the agents for control signal computation and after agent operation in JADE the necessary control signals are sent back to the actuators in Matlab. Several simulation scenarios were performed to demonstrate the feasibility and the possible benefits of this control architecture.