Search

• Data-driven dynamic harmonic coupled admittance matrix model is proposed for modern home appliances. • Recursive partial least squares method and the model adaptive update strategy are used for Characterizing the dynamic harmonics. • Harmonic voltage evaluation, line power loss calculation including harmonics and energy savings are realized by using the dynamic model. With the rapid development of power electronics technology and the enhancement of users' energy conservation awareness, the proportion of modern household appliances with power electronic devices is increasing year by year, which has become an important harmonic source. Together with the massive access of home photovoltaic and electric vehicles, the harmonics of the existing residential low-voltage distribution system (RLVDS) show higher distortions and stronger dynamics than ever before. Therefore, an accurate harmonic model of the modern household appliance is necessary for power quality analysis and improvement. This paper proposes a data-driven dynamic harmonic coupled admittance matrix model (HCAMM) based on measured voltage and current data. First, the parameters of the HCAMM are solved by the partial least squares method (PLSM), which can effectively overcome the matrix pathology problem in the solution process, and the HCAMM can reflect the coupling effect between each harmonic. Second, to address the voltage dynamics, the dynamic HCAMM is established by combining the recursive PLSM and the adaptive update strategy, which can accurately characterize the harmonics under different supply voltages. Finally, the accuracy, dynamics, and generality of the proposed harmonic model are verified by simulations and experiments, and the harmonic model is applied to the power quality harmonic evaluation, line power loss calculation with harmonics, and energy saving by reducing the harmonics. [ABSTRACT FROM AUTHOR]

Solar energy is one of the most promising energy types, but photovoltaic (PV) mismatches have several negative effects. Finding an efficient and low-cost method to reduce or eliminate the power loss caused by the mismatch in PV power generation systems is necessary. In this paper, we propose a PV submodule-distributed maximum power point tracking (DMPPT) scheme based on a multi-port DC converter. In-depth analysis and research are performed on the PV model and mismatch characteristics, the PV module voltage equalization (VE) strategy, and implementation with the multi-port buck–boost converter. We discuss the trade-off between true DMPPT and VE on control overhead and tracking accuracy, and a detailed derivation is created to illustrate the differences between them. The VE can reduce the cost and complexity of the system at the expense of a small PV energy capture. The proposed implementation reduces the number of semiconductor components and current sensors, which is conducive to the integrated design of the hardware. A prototype based on a multi-port buck–boost converter for a 36 V/200 W PV module was designed and built. The feasibility and progress of the proposed scheme were verified through experiments. [ABSTRACT FROM AUTHOR]