A system reliability trade-off study for centralized and distributed PV architectures

In central inverter based photovoltaic (PV) plants, critical functions such as maximum power point tracking (MPPT) are performed on the aggregate I-V curve of all of the PV modules. The disadvantage of this approach is its inability to optimize overall output power when the operating conditions vary from one PV module to another, due to module differences (e.g. mismatch, layout, ...) and shading effects among other things. The use of DC/DC (e.g. optimizers) and DC/AC (e.g. microinverters, string inverters) converters either at the module or at the string level addresses this problem by enabling localized, or distributed maximum power tracking. In this paper, we report on the results of a simulation-based reliability analysis of a centralized versus distributed DC/DC converter PV architecture. Our analysis shows that a PV architecture with distributed MPPT improves overall system reliability and availability compared to the central inverter architecture.