Practical Terminal Sliding Mode Control of Active Power Filters by Self-Organizing Hermite Fuzzy Neural Network

To mitigate harmonic contamination, this study introduces a super-twisting practical terminal sliding mode controller (STPTSMC) using a self-organizing Hermite fuzzy neural network (SOHFNN) for an active power filter (APF) with unknown model nonlinearities. First, a practical terminal sliding mode (PTSM) because of its benefits in offering a continuous and nonsingular characteristic is adopted, and a super-twisting sliding mode control (STSMC) algorithm is embedded to smoothen chattering control signals and enhance APF's dynamic tracking capability. Second, the fusion of fuzzy neural network, self-organizing mechanism, feature selection algorithm, and Hermite activation function empowers the SOHFNN to precisely estimate the uncertain STPTSMC owing to component parameter perturbation of APF. Finally, the feasibility and efficacy of the SOHFNN-STPTSMC scheme have been validated through simulation and experimental efforts, evidencing its improvement performance relative to other methodologies.