Comparative performance analysis of adaptive neuro-fuzzy inference system, fuzzy logic, and neural network-based maximum power point tracking controllers for permanent magnet synchronous generator-based variable-speed wind energy conversion systems

In this study, the performance of MPPT based Fuzzy Logic Controllers (FLC), Artificial Neural Networks (ANN) and Adaptive Neuro-Fuzzy Inference Systems (ANFIS) for a permanent magnet synchronous generator (PMSG) based variable speed wind energy conversion system (WECS) is evaluated. Controllers are designed in MATLAB/Simulink and subjected to stepwise and time-varying wind speed settings to determine mechanical power extraction, boosted peak voltage, transient response, voltage regulation and steady-state stability. The stepwise wind speed is changed from 5 to 12 m/s and then to 9 and 10 m/s to assess the controller performance under dynamic operating conditions. The results show that all three intelligent controllers are able to successfully track the maximum power operating point, extracting 16.8 kW under the examined conditions, and get a maximum power coefficient of \(C_p=0.48\) . But large differences are observed in their electrical and dynamic responses. ANFIS achieves the best overall performance with maximum boosted voltage of 600 V and steady state DC-link voltage of 480 V with minimum oscillation. It also has the best damping properties, fastest transient response and least settling time compared with FLC and ANN. FLC provides intermediate dynamic performance with less computing complexity. ANN has relatively strong transient oscillations and longest settling response. The results demonstrate that ANFIS offers better dynamic stability and voltage regulation while all three controllers deliver similar maximum power extraction. Thus, ANFIS is proposed as a promising intelligent MPPT technique to enhance the performance and reliability of PMSG based variable-speed WECS under fast varying wind conditions.