Intelligent GA and PSO-Tuned Controllers for High-Performance Direct Field-Oriented Control in DFIG-Based WECS

Doubly Fed Induction Generators (DFIGs) have been frequently controlled through Direct Field-Oriented Control (DFOC) in Wind Energy Conversion Systems due to their ability to independently regulate both active and reactive power, but their performance critically depends on the tuning of PI controller parameters. Manually tuned PI controllers often lead to degraded power quality under varying conditions. To address this challenge, this paper proposes two intelligent control approaches, DFOC-GA-PI and DFOC-PSO-PI, incorporating Genetic Algorithm (GA) and Particle Swarm Optimization (PSO), respectively, to tune DFOC’s power controller. The suggested approaches adjust controllers by utilizing a multi-objective fitness function that considers multiple performance metrics. Comprehensive Matlab/Simulink simulations were performed under realistic conditions to demonstrate that DFOC-GA-PI notably outperforms traditional DFOC (DFOC-C-PI), delivering a 46% enhancement in active power fluctuation, 57% enhancement in THD, and 25% faster response time. DFOC-PSO-PI achieves faster convergence compared to DFOC-GA-PI; however, it exhibits slightly lower performance. The robustness assessment under realistic varying wind and DFIG parameters proves the resilience of DFOC-GA-PI frameworks against parameter variations. Comparative examinations under three different scenarios, including nominal, moderate, and the most challenging conditions, demonstrate the superiority of both proposed DFOCs in improving power quality, THD of current, and overall performance, ensuring excellent grid compliance. Depending on the populations/particles size and generations/iterations, the developed DFOC frameworks exhibit moderate computational complexity, and convergence is achieved within approximately 4 seconds on a standard PC (intel i7/ 16 RAM), which can be reduced to a few microseconds on modern calculators or high-performance controllers, confirming the practical feasibility of these developed frameworks.