一种用于光伏电池等效电路模型参数在线估计的直接连续时间系统辨识方法

The AC characterization of solar photovoltaic (PV) cells typically employs electrical impedance spectroscopy (EIS) to ascertain internal dynamic impedance. However, EIS faces limitations in real-world applications, particularly under varying environmental conditions. The standard EIS method uses perturbation-based techniques tailored for specific frequency bandwidth measurements. To mitigate these issues, this paper introduces a direct continuous-time (CT) system identification strategy based on the recursive instrumental variable approach to estimate the AC equivalent electrical circuit (EEC) parameters of the solar PV. This CT identification method outperforms the EIS frequency technique in estimating AC EEC parameters under changing environments. It exploits the DC-DC boost converter's inherent switching characteristics, eliminating the need for an external excitation signal. The focus lies on the recursive instrumental variable method implemented using the Kalman filter. Simulation and experimental results validate the analysis and demonstrate the advantages of the CT system identification method in solar PV AC characterization applications. In the simulation, the EIS procedure and the system identification algorithm were compared to ground truth values. A Frequency Response Analyzer (FRA) was utilized to benchmark the experimental results estimated using the system identification method.