Artificial-Neural-Network-Based Phase-Locking Scheme for Active Power Filters

Abstract—Artificial-Neural-Network-Based Phase-Locking Scheme for Active Power Filters. This paper presents a phase-locking control scheme based on artificial neural networks (ANNs) for active power filters (APFs). The proposed phase locking is achieved by estimating the fundamental supply frequency and by generating a phase-locking signal. The nonlinear-least-squares-based approach is modified to estimate the supply frequency. To improve the accuracy of frequency estimation, < Final Year Projects > when the supply voltage contains harmonics that are not known, a prefiltering stage is introduced. In shunt APF applications, not only the information of frequency is sufficient but also the phase information of the supply voltage is required to generate a unit template that is phase-locked to the supply voltage. Therefore, in this paper, an adaptive-linear-neuron-based scheme is proposed to extract the phase information of the supply voltage. The estimated system frequency and phase information are then utilized to generate a phase-locking signal that assures a perfect synchronization with the fundamental supply voltage. To demonstrate the effectiveness of the proposed approach, the synchronous reference frame ( d-q theory) shunt APF control method with the proposed ANN-based phase-locking scheme is adopted. The performance of the proposed ANN-based approach is verified experimentally with different supply systems and load conditions.