An Advanced Shunt Active Power Filter (SAPF) for Non-ideal Grid Using Predictive DPC

The usage of power electronic converters and other non-linear devices is very common in this modern era because of their flexibility in control. However, they inject harmonics into the power system. To deal with the problem, the use of filters has become almost inevitable. Active power filters controlled by direct power control (DPC) have overtaken this responsibility from the passive filters because of their numerous advantages. To deal with the real-world power system grid which is far from the idealized character of being balanced and sinusoidal, the proposed method uses the extended pq theory integrated with the dual second-order generalized integrator (D-SOGI) for the correct estimation of the all-important instantaneous powers. In contrast, the conventional pq theory is used in the existing works, which doesn't hold in a non-ideal grid. In addition to that, here, for the selection of the switching vector, an advanced and more accurate predictive DPC method which is specifically designed to operate in a non-ideal grid, replaces the static look-up tables used in the previous works. Further, to make the filter designing easier, the proposed method focuses on constant switching frequency operation eliminating the problems arising from variable switching frequency used in literature. The proposed technique is tested in MATLAB/Simulink environment and the real-time system to validate its feasibility.