Guardat en:
Dades bibliogràfiques
Autor principal: Souza, Vitor Amadeu
Format: Recurso digital
Idioma:
Publicat: Zenodo 2026
Matèries:
Accés en línia:https://doi.org/10.5281/zenodo.19610313
Etiquetes: Afegir etiqueta
Sense etiquetes, Sigues el primer a etiquetar aquest registre!
Taula de continguts:
  • <div class="markdown markdown-main-panel enable-updated-hr-color"> <p>This article presents the development of a virtual sensor for meteorological variables implemented in Python, with real-time data collection via the OpenWeatherMap API, sliding-window in-memory storage, and short-term forecasting of temperature and relative humidity through linear regression. The system architecture integrates a Flask web server, a machine learning module based on the Scikit-learn library, and an interactive graphical interface built with Chart.js, which displays the observed historical record alongside model-generated forecasts. The results demonstrate that the system is capable of continuously monitoring variables such as temperature, relative humidity, atmospheric pressure, and wind speed, producing short-term forecasts based on recent trends. The proposed approach is computationally lightweight, straightforward to deploy, and constitutes a viable alternative for real-time environmental monitoring applications operating under limited computational resources.</p> </div>