Performance analysis of bilinear and volterra harmonic estimator for microgrid applications
摘要
Harmonic estimation is a crucial challenge in the current microgrid/smart grid-based system, as integrating renewable sources into the grid injects multiple orders of harmonics into the system. Thus, the amplitude and phase parameters of the harmonic components need to be estimated correctly. Adaptive filtering-based models efficiently track the time-varying harmonics present in the system. In this paper, a detailed comparative analysis is made using Volterra and bilinear adaptive estimation models while tracking the amplitude and phase parameters. The main idea behind the development of efficient harmonic estimator is based on the integration of Volterra and Bilinear expansion blocks to the objective functions of LMS and LMS/F algorithms. Most importantly, the distorted voltage signals generated from the single-phase inverter model, the PV-based model, and an IEEE voltage signal dataset are used to observe the harmonic estimation accuracy of different models. The quantitative performance comparison is made with other least LMS families of estimation models by calculating THD and MSE. A separate section is included to explain the FPGA modeling of the harmonic estimator by integrating the hardware board.