{"ok":true,"nota":{"id":"d26e56578791ff75","slug":"un-sistema-con-inteligencia-artificial-predice-la-temperatura-y-acidez-d26e56","titulo":"Un sistema con inteligencia artificial predice la temperatura y acidez del agua de riego en invernaderos de Jo","resumen":"Un equipo desarrolló modelos que anticipan con 12 horas de antelación la conductividad y el pH del agua de drenaje en invernaderos, y también la producción semanal. El sistema corre en una Raspberry Pi y envía los datos a la nube para visualizarlos en un panel.","fuente":"zenodo","fuenteNombre":"Zenodo · Investigación abierta","color":"#1682d4","autor":"Hindi, Iman, AlZaareer, Khaled","url":"https://doi.org/10.5281/zenodo.21645524","imagen":"/hoy/portada/d26e56578791ff75","tema":"ciencia","subtemas":["agricultura inteligente","riego","invernaderos","inteligencia artificial","agua escasa"],"entidades":["Jordania"],"formato":"dato","entretenimiento":false,"pubTs":"2026-10-20T03:03:03.000Z","indexTs":"2026-10-22T22:31:26.228Z","hash":"42f2d7dc816b00ea167b4aab78a6a3c39793a5e74da72d0c598ca5d9c879aaa6","tsa":{"time":"Oct 22 22:32:23 2026 GMT","provider":"freetsa.org","sobre":"sello"},"stats":{"views":0,"likes":0,"clicks":0,"dwell":0},"tipo":"normal","registro":"hallazgo","porQueImporta":"En zonas donde el agua escasea, saber con horas de anticipación cuánta sal y qué acidez tiene el agua de riego permite ajustar la fertilización sin desperdiciar nada. Un productor con un equipo barato podría usar una herramienta así para decidir mejor cuándo regar y cuánto fertilizante aplicar. El trabajo está pensado como apoyo a decisiones, no como reemplazo del criterio de quien trabaja la tierra, y los datos de entrenamiento no se publican completos.","quienFirma":"Estudio sobre agricultura inteligente en Jordania (Parte II); institución no especificada en la fuente","confianza":"revisado","consulta":null},"sello":{"hash":"42f2d7dc816b00ea167b4aab78a6a3c39793a5e74da72d0c598ca5d9c879aaa6","sello":"925b08b6dc6d9a81932c88a0742239a17f4f71a4ac00dea2e92815d82be2517f","beacon":{"drand":{"cadena":"8990e7a9aaed2ffed73dbd7092123d6f289930540d7651336225dc172e51b2ce","ronda":6575910,"aleatorio":"ad632c470ba8c968ea920b116993e6a501688139a43318de64f5a7ea72e81de2","tiempo":"2026-10-22T22:32:00.000Z"},"nist":{"cadena":2,"pulso":1995154,"salida":"A32727015042185E787530FED53F828C26131B55C644F3F364155FFBAB02AAA2930B78FFEE6913417A86FA34480E1F05FB9B4984B59D1A0E007BB8E6E3A95F8F","tiempo":"2026-10-22T22:31:00.000Z"},"noAntes":"2026-10-22T22:32:00.000Z"},"verificacion":{"drand":true,"nist":true},"tsa":{"time":"Oct 22 22:32:23 2026 GMT","provider":"freetsa.org","sobre":"sello"},"original":{"titulo":"Leakage-Controlled Cross-Compartment Forecasting of Harvest Production and 12-Hour EC/pH Dynamics","descripcion":"<p>This software repository accompanies Part II of the study on smart and sustainable agriculture in water-scarce Jordan. It provides a reproducible workflow for greenhouse data preprocessing, 12-hour drain EC and pH forecasting, weekly production prediction, model evaluation and explainability, Raspberry Pi edge deployment, LoRa point-to-point data reception, ThingSpeak cloud integration, and Power BI visualization.</p> <p>The archive includes the training notebook, fitted model artifacts, inference and replay scripts, selected reproducibility outputs, configuration templates, dashboard files, and documentation. The complete Autonomous Greenhouse Challenge dataset is not redistributed; authorized access and expected data organization are described in the data-availability documentation.</p> <p>The models are intended for research and decision support. Application outside the training do","pubDate":"2026-10-20T03:03:03Z","url":"https://doi.org/10.5281/zenodo.21645524"},"indexado":"2026-10-22T22:31:26.228Z"}}