An adaptive Zhang Neural Network controller for Frequency Control of Renewable Energy Integrated System

This paper proposes a Zhang neural network (ZNN) designed self-adaptive proportional-integral-derivative (PID) controller for frequency control of renewable energy integrated systems. The network is formulated to minimize the error function that minimizes the area control error of the integrated system by optimizing the controller. Initially, the control problem is formulated as an error function in terms of area control error associated with gains of PID controller such as K
p
, K
i
, and K
d
. Then, the gradient equations governing the dynamics of Zhang Gradients (ZG) are derived from the error function. The presented method is simulated in MATLAB/Simulink and the results obtained have shown the ZNN-based PID controller gives a smooth and faster response than simple ZG and Hopfield neural network-based PID controllers. To validate the robustness of the controller, the system is tested in the presence of random load disturbance, and the performance of the proposed controller is more predominant. In the case of consecutive changes in load demand, the values of K
p
, K
i
, and K
d
are adapted with respect to the plant dynamics, demonstrating the self-adaptiveness of the controller.