Unit-Minimum Least Power Point Tracking for the Optimization of Photovoltaic Differential Power Processing Systems

Abstract

Recently, the module integrated converter and differential power processing (DPP) architecture were introduced to enable the photovoltaic (PV) power conditioning system to maintain the optimal operating condition of PV cells, such as maximum power point tracking (MPPT), even under partial shading conditions. However, the DPP architecture was found to have more room to optimize the performance of the systems, by the application of an extra extremum-seeking control, the so-called least power point tracking (LPPT) method that was introduced last year. The main idea is that most of the power from the PV modules is processed through the main-string high-efficiency nonisolation converter, and only a minimal fraction of the power that changes depending on the PV-string current is transferred through the low-efficiency bidirectional isolated DPP converters. This paper suggests a second version of the LPPT method, called unit-minimum power-distributing LPPT, which improves the first version of LPPT, called total-minimum centralized power-distributing LPPT. Instead of minimizing the total power of DPPs, the proposed LPPT minimizes the power of the DPP converter unit, which is the largest among them. Then, the system size and cost can be reduced by the proposed LPPT method, which enables the multiple DPP converters to have smaller power capacity and losses than those of the previous LPPT method. The real-time extremum-seeking algorithm employs a perturb-and-observe method, which comes from the conventional MPPT one, while the optimization process directs minimal extremity, not maximal. The peak system efficiency achieved with a 400-W prototype DPP system employing the LPPT algorithm is 96.7%.