Least Power Point Tracking Method for Photovoltaic Differential Power Processing Systems

Abstract

Differential power processing (DPP) systems are a promising architecture for future photovoltaic (PV) power systems that achieve high system efficiency through processing a faction of the full PV power, while achieving distributed local maximum power point tracking (MPPT). In the PV-to-bus DPP architecture, the power processed through the DPP converters depends on the string current, which must be controlled to minimize the power processed through the DPP converters. A real-time least power point tracking (LPPT) method is proposed to minimize power stress on PV DPP converters. Mathematical analysis shows the uniqueness of the least power point for the total power processed through the system. The perturb-and-observe LPPT method is presented that enables the DPP converters to maintain optimal operating conditions, while reducing the total power loss and converter stress. This work validates through simulation and experimentation that LPPT in the string-level converter successfully operates with MPPT in the DPP converters to maximize output power for the PV-to-bus architecture. Hardware prototypes were developed and tested at 140 and 300 W, and the LPPT control algorithm showed effective operation under steady-state operation and an irradiance step change. Peak system efficiency achieved with a 140-W prototype DPP system employing LPPT is 95.7%.