Frequency control of renewable energy integrated power systems using artificial intelligence based advanced controllers

Abstract

newline This dissertation primarily focuses on the area of hybrid power system stability advancement by employing reinforced recent Artificial Intelligence (AI) techniques considering various critical operational nonlinearities associated with Load frequency control (LFC) for providing reliable and secure power from the utility grid. A maiden attempt has been taken for the frequency stability improvement of microgrid and hybrid power systems by introducing the latest improved and hybridized soft computing techniques with various advanced adaptive fuzzy based control approaches. Among various soft computing techniques, meta-heuristic approaches such as improved Moth Flame Optimization (I-MFO) algorithm, teaching learning-based optimization (TLBO) algorithm, Improved Grey Wolf optimization (I-GWO) algorithm, Improved Equilibrium optimization algorithm (i-EOA) are more effective in LFC optimization problems as compared to traditional Genetic algorithm (GA), Particle Swarm optimization (PSO) algorithm, Grey Wolf optimization (GWO) algorithm, Moth Flame Optimization (MFO) algorithm and Equilibrium optimization algorithm (EOA) and hence these meta-heuristic approaches are implemented in considered Microgrid and hybrid power system networks. Due to the active participation of Renewable energy sources in the utility grid, the system complications are enhanced in the modern power system networks. Therefore, robust, smart and intelligent control mechanisms are essential to be deployed in the power system network for the advancement of frequency stability. So, control schemes such as Tilt multistage TDF/(1+TI) controller, Fractional order Fuzzy PID (Fuzzy FOPID) controller,Trapezoidal membership accessed fuzzy type-2 controller (TM-T2FC), Fuzzy Tilted Double Integral Derivative with Filter (F-TIDF-2) Controller are introduced which advances the dynamic performances of frequency regulation efficiently as compared to classical controllers such as Proportional-Integral-Derivative (PID) controller, multistage PD/1+PI, con