Grid voltage estimation based on an adaptive linear neural network for pv-Active power filter control strategy

To enhance the power quality in electric power systems, active power filter can be successfully used for harmonics mitigation and reactive power compensation. In this paper, a new approach using novel direct power control PDPC based on grid voltages estimation is proposed to improve the dynamic performance of shunt active power filter and reduce the system cost and robustness by minimizing the number of sensors in grid connected PV system. The grid voltages are estimated online by using an adaptive linear neural network (ADALINE) and the proposed PDPC is based on the extended pq theory. According to the simulation results, the proposed control strategy using a voltage estimator instead of voltage sensors not only reduce the size and cost but the reliability of the active power filter PV-APF is improved under normal and severe grid voltage conditions. © 2021 Lavoisier. All rights reserved.