Green anaconda optimized DRN controller for automatic generation control of two-area interconnected wind–solar–tidal system.

In this work, two area systems of renewable energy sources (RES), such as tidal, wind and solar, are interconnected with the deep residual network (DRN) controller in a microgrid. The DRN controller is used to control load frequency, which is optimally designed by green anaconda optimization (GAO) algorithm. The GAO algorithm is used to optimally select the DRN layers that improve the control of a two-area interconnected power system. GAO algorithm is utilized for the parameter section of a DRN controller in the interconnected power system with two areas, which provides optimal solutions based on the natural behaviour of green anacondas. The reason for choosing the GAO method is that it has high capability in the exploration and exploitation phase, thus creating effective tuning of controllers. The DRN controller is chosen because it has fewer additional parameters and increases the accuracy level, thus enhancing control of the two areas. The proposed model is implemented in MATLAB tool, and the results are compared with the existing methods of proportional-integral derivative-whale optimization algorithm (PID-WOA), PID-Harris hawks optimization (PID-HHO) and two degrees of freedom (2DOF)-PID (2DOF-PID). The comparison results indicate that the proposed GAO-based DRN controller achieves 98% efficiency and 95% frequency deviation, thus providing high-performance over counterpart methods. [ABSTRACT FROM AUTHOR]