Fast and accurate estimation of multifrequency signals with noise and harmonics in control of photovoltaic systems

Abstract

Photovoltaic systems are the important part of the energy market in many countries around the world. Such systems contain several parts and one of them is the inverter to produce the AC signal. In this paper, an estimation method is used to obtain the value of frequency, amplitude and phase of the grid signal used to drive the inverter in a proper way. The method is based on the 3-point spectrum interpolation and I class Rife-Vincent time windows. The tested signal contains harmonics and AWGN noise to reflect real and practical measurement conditions. Additionally, simulations were performed for various values of measurement parameters to show broadly properties of the method. The estimation time is only up to two CiR (cycles of the signal in the measurement window). Harmonic components and noise negatively affect the accuracy but the method is so flexible that it can be adjusted to the needs of a current estimation. The estimation error for SNR = 50–60 dB and for two additional harmonics is in the order of approximately 10^−3.

Keywords

• interpolation
• awgn
• photovoltaic power systems
• power system harmonics
• frequency estimation
• power generation control
• power markets
• power generation economics
• amplitude estimation
• phase estimation
• power grids
• invertors
• energy market
• inverter
• 3-point spectrum interpolation
• awgn noise
• current estimation
• photovoltaic control systems
• multifrequency signal estimation error method
• power system harmonic components
• measurement window parameters
• amplitude estimation
• phase estimation
• i class rife-vincent time windows
• noise figure 50.0 db to 60.0 db