Inverter Output Impedance Estimation in Power Networks: A Variable Direction Forgetting Recursive-Least-Square Algorithm Based Approach

Jaesang Park,Alireza Askarian,Srinivasa Salapaka

2024-10-18

Abstract:As inverter-based loads and energy sources become increasingly prevalent, accurate line impedance estimation between inverters and the grid is essential for optimizing performance and enhancing control strategies. This paper presents a non-invasive estimation algorithm that avoids signal injection, based on the Variable Direction Forgetting Recursive Least Squares (VDF-RLS) method. The method uses measurement data that is local to the inverter. It proposes a specific method for determining rotational frequency for direct-quadrature (dq) coordinate frame in which data is collected, which ensures a simpler and more accurate estimation. This method is enabled by a secondary Phase Locked Loop (PLL) which appropriately attenuates the effects of variations in grid-voltage measurements. By isolating the variation-sensitive q-axis and relying solely on the less sensitive d-axis, the method further minimizes the impact of variations. The estimation method achieves rapid adaptation while ensuring stability in the absence of persistent excitation by selectively discarding outdated data during updates. Results demonstrate significant improvement (as large as 7 times) in estimation of line parameters, when compared to existing approaches such as constant forgetting RLS.

Systems and Control