Authors : Isra`a L. Salim, Osama Ali Awad and C.T. Kalaivani
Abstract: Networked Control System (NCS) has recently been a hotspot in the research fields of control theory and control engineering application where control system components (controller and controlled system or plant) are connected via a shared network. Taking a third-order DC motor as the plant and using the true time toolbox in MATLAB, a NCS Model is designed with Ethernet network as the communication channel. Through the simulation, different traffic scenarios on Ethernet network are studied. Different load conditions induce timevarying delays between measurements and control. To face these variations a delay-dependent gain scheduling Proportional-Integral-Derivative (PID) controller has been designed a hybrid technique for gain scheduling of the PID controller time delay system. The proposed hybrid technique will be the combination of the Artificial Bee Colony (ABC) algorithm and Artificial Intelligence (AI) technique. The ABC algorithm will be optimized the gain parameters and generates the training dataset based on the variation of the output response of the system. Here, the ABC utilizes least square error is the objective function. Based on the training dataset the AI technique will be trained and the gain parameters of the PID controller will be scheduled. By using the proposed technique, the PID controller will be tuned and the stability of the system will be maintained. The ABC technique is one of the non linear programming techniques which can constructed that approximately fits the data from the available solution space. So, the fit poles, the optimal controller gains and parameters will be determined to the system. The proposed hybrid technique is implemented in MATLAB working platform and the tuning performance is evaluated.
Isra`a L. Salim, Osama Ali Awad and C.T. Kalaivani, 2016. Gain Scheduling of the PID Controller Time Delay System Using Hybrid Methodology for Network Control System. Asian Journal of Information Technology, 15: 5075-5083.