Optimal PI Pitch Control of SCIG Wind Turbine Using Grey Wolf Optimizer for Dynamic Stability

The optimal tuning of Proportional Integral (PI) pitch control of the Wind Turbine is necessary for optimal performance in power limitation to rated value and reducing loading on wind turbine components. Traditionally, the Ziegler Nichols (ZN) tuning method is applied to tune the PI controller for pitch angle control. Nowadays, the soft computation meta-heuristic optimisation techniques are used to fine-tune the PI controllers for precise and optimal control of Wind Turbines (WT). In this study, the Grey Wolf Optimization (GWO) technique is proposed to tune the PI controller for the pitch angle control of the SCIG-WT. The objective is to minimize the Integral Time multiplied Square Error (ITSE) objective function based on the power error of the WT using GWO. This ensures the best combination of optimised PI parameters of the pitch control and hence the optimised dynamic response of the SCIG WT operating in power limitation mode under disturbance such as the sudden high wind speed and change of air density in bad weather conditions. To validate the proposed GWO tuning technique, it is performance is compared with a meta-heuristic optimisation technique, namely the canonical Particle Swarm Optimization (PSO), the evolution-based Genetic Algorithm (GA) and the Ziegler Nichols (ZN) tuning method. The GWO shows its supremacy and effectiveness in terms of fast convergence of tuning characteristics compared to other algorithms. Furthermore, from sudden change above rated wind speed or air density in the SCIG-WT, the GWO tuned PI controller pitch control provided the least settling time and overshoot in pitch angle compared to PSO, GA and ZN tuned PI controllers. In addition, it provided the least undershoot of the WT active power and rotor speed compared to PSO, GA and ZN tuned PI controllers. Therefore, it can be concluded that GWO tuned PI controller implemented in pitch control provided better dynamic stability of SCIG-WT output compared to PSO, GA, and ZN tuned PI controllers.