Resumen: This paper presents an in-depth analysis and simulation of the effects produced by cloud passages on the operation of a stand-alone direct photovoltaic (PV) water pumping system. The study is based on a long-term experimental dataset collected at an isolated PV installation in Cariñena (Spain), specifically designed for agricultural irrigation. Cloud events were classified into three main types -cloud trains, compact clouds, and isolated cloud events- and their statistical occurrence and impact on system performance were evaluated across different seasons. A data-logger was developed to monitor key electrical and environmental parameters with high temporal resolution, enabling the identification and quantification of transient effects associated with sudden irradiance drops. The most critical situation arises during sequences of cloud trains, where frequent restarts and short operating cycles reduce the overall efficiency and increase mechanical stress. A simulation model was implemented to reproduce the dynamic behaviour of the system during cloud events, validating the experimental results and offering predictive capabilities for performance estimation. The novelty of the approach lies in the integration of cloud passage statistics with real-time performance analysis, allowing for a deeper understanding of the mismatch between energy generation and hydraulic demand. The study also highlights the advantages of direct PV pumping systems, such as simplicity, low maintenance, and cost-effectiveness, while highlighting their limitations under unstable weather conditions. Idioma: Inglés DOI: 10.24084/reepqj25-435 Año: 2025 Publicado en: Renewable Energies, Environment & Power Quality Journal 3 (2025), 273-281 ISSN: 3020-531X Financiación: info:eu-repo/grantAgreement/EUR/AEI/TED2021-129801B-I00 Tipo y forma: Article (Published version) Área (Departamento): Área Ingeniería Eléctrica (Dpto. Ingeniería Eléctrica)