Hardware in the loop implementation of the wideband identification of three phase AC power grid impedances using an existing grid-tied power electronics inverter

In smart grid applications, online wideband monitoring of AC power grid impedances is a key enabler of a set of capabilities, such as monitoring of stability margins, active filter tuning, and adaptive control of inverters. In order to realize all these smart grid applications, this work here proposes the implementation of an online wideband system identification technique, via Hardware In the Loop (HIL) real-time simulation. The impedance identification technique uses an existing grid-tied inverter for the estimation of wide bandwidth three-phase AC grid impedance. Recently people are more interested wideband measurements of AC power system impedances. The AC power distribution system has an increase of nonlinear, harmonic-producing loads and distributed sources interfaced via power electronic converters with high switching frequency. As a result, the distribution power system is becoming a wideband network. However, since many of the converters already connected to the AC grid have digital control and sensing, they may also be used to monitor system dynamic responses and stability margins. Digital network analyzer techniques can be integrated into the converter controllers allowing them to be used as online monitors without any extra hardware. This thesis focuses on the identification of three-phase grid impedances via an existing grid-connected inverter. The technique is to inject a small amplitude white noise signal to be superimposed on the duty cycle of the inverter and then estimate the wideband impedance from the voltage and current measurements over the length of the injection. In particular, the perturbation consists of a Pseudo Random Binary Sequence (PRBS), which gives a small-amplitude disturbance into the grid that is wide bandwidth in nature, so that all frequencies of interest can be excited simultaneously. Cross correlation techniques are then applied to extract the impedance from the measured response to the perturbation. The identification method is to implement in MATLAB/Simulink as a preliminary result, and a Hardware In the Loop (HIL) real-time simulation setup will be presented for the identification as the second step. The HIL setup will be consisting of the existing grid-tied inverter simulated in OPAL-RT and the hardware under test where the identification algorithms are implemented. This setup, can achieve the following unique outcomes that a simple real-time simulation setup cannot show: the possibility of the online identification of real-time simulated grid impedances and the numerical evaluation of the identification processing performance.