Comparative performance analysis of direct-coupled DC and variable-frequency AC photovoltaic water pumping systems under tropical climatic conditions

Abstract Rural tropical areas often lack grid access and increasingly rely on photovoltaic water pumping. This study numerically compares two architectures under identical conditions (PV array, motor pump, hydraulic system, and demand): a direct-coupled DC system and a variable-frequency AC system with MPPT. Simulations use Typical Meteorological Year data (2005–2023) from four Cameroon sites, Meskine, Agola, Mbé, and Soa, at a 1-min time step. The PV generator is modeled with a single-diode model, the DC branch includes electromechanical motor dynamics, while the AC branch uses a quasi-steady power transfer model with an adaptive perturb-and-observe MPPT. Results show that the AC annual pumped volume exceeds DC at three sites: +18.9% at Meskine, + 24.3% at Agola, and + 7.8% at Mbé. At Soa, no AC gain was observed (− 1.8%, not significant). This ranking follows the local climate and is mirrored in energy efficiency, temperature-corrected Performance Ratio, and operating time. Decomposing the AC gain shows that it stems mainly from better utilization of photovoltaic power through MPPT, which captures 84–87% of the theoretical potential at high-insolation sites versus 66% at Mbé, rather than from a faster restart response at the temporal resolution considered. AC superiority is therefore not universal but climate-dependent, which argues for a site-specific assessment before any technological choice is made. An explicit limitation of this study is that the irradiance time series, available hourly and then extrapolated to one-minute steps, cannot resolve sub-hourly transients.