Modified window, recursive least square estimator for active and reactive powers in single-phase AC systems

Abstract

In this paper, an Instantaneous Power Signal (IPS) has been used to extract amount of active/reactive power in real-time. Estimation of real-time power is challenging in single-phase systems due to the presence of double-frequency ripple in IPS. In conventional methods the low bandwidth filters are used to remove this ripple. In this paper the real-time power estimation is obtained first by presenting the problem in terms time dependent linear optimization and then solved by using a least square (LS) method. In order to tackle the drawbacks associated with the conventional methods based on dq rotating frames and grid synchronization algorithms, the proposed algorithm is modified to be recursive and windows type to reduce the computational time. It is mathematically shown that by adding one additional parameter the active and reactive power estimation can be made independent of the initial time value. Simulation and experimental results prove the superior performance of the proposed estimator in comparison with the conventional estimators and the other least square algorithms, such as Kalman filters and forgetting factor, in term of convergence time and the amount of undershoot/overshoot in estimation of active and reactive power from instantaneous power signal.