Sistema de Riego Inteligente Basado en IA para la Optimización del Uso de Agua en Cafetales

Abstract

Abstract: This paper presents the design and validation of a smart irrigation system to improve water management in Andean coffee plantations. The proposal was developed in a simulation environment, integrating digital tools based on the Industry 4.0 approach. The infrastructure was designed in layers to organize the capture, transmission, and analysis of environmental information; to this end, sensors were used and communication was established via MQTT, complementing the system with a time-series database for storing the generated data. Additionally, climate data from NASA POWER was incorporated, and soil moisture was modeled in a simplified manner based on precipitation and temperature. The control scheme was implemented using deterministic rules based on predefined thresholds, which allowed for the activation or suspension of irrigation based on environmental conditions. Validation was performed by analyzing scenarios involving water deficit, rainfall, and soil saturation.