A Robust Lyapunov’s Demodulator for Tracking of Single-/Three-Phase Grid Voltage Variables

This article proposes a robust Lyapunov's demodulator (LD) based on an orthogonal signal generation (OSG) approach for single-/three-phase applications. The LD is not normally capable of rejecting the dc-offset in the grid signal, and an additional estimation loop is, therefore, required, which may affect the dynamic performance. Nevertheless, an application of a harmonically polluted grid voltage signal to the LD may severely affect the steady-state performance of the parameters estimated. These issues are addressed by proposing an enhanced LD-based-OSG in which a moving average filter is incorporated into the LD-OSG structure. The rapid rejection of dc-offset and harmonics is, therefore, easily achieved without any additional loop. The proposed structure can accurately estimate the fundamental in-phase and quadrature components. However, these orthogonal components may suffer from amplitude imbalance and errors in the phase information under off-nominal frequency conditions. Nevertheless, the errors in the amplitude and the phase information are eliminated using an open-loop frequency deviation detector and a feed-forward curve-fitting approach. The dynamic performance of the proposed scheme has been validated by means of numerical and hardware studies whose results show that, with the less sensitivity toward the phase angle jump and good immunity to the fundamental negative sequence, the proposed scheme is a potential technique with which to synchronize single-/three-phase grid-connected power electronic equipment.

I. Introduction

The evolution of the smart grid has led to a situation in which any improvements to grid resilience and power quality rely on a good control over grid-connected converters (GCCs) [1], [2]. The control architecture of GCCs is composed of well-known grid synchronization algorithms, i.e., phase-locked–loops (PLLs) and frequency locked–loops [3]–[14]. The high performance of a control algorithm is generally achieved by means of analog or digital signal conditioning units. With the pertinent digital signal processors, the robust performance of a control technique relies on the rapid removal of the noise and the harmonic components [4], and there is consequently a need for research into immunity-based digital filtering techniques [5]. The optimized filtering techniques often applied in order to develop better control architecture of GCCs are reported in [6]–[8].