Scholarlink

<p>Online weather forecast system usually works over a large geographical area. However, it has been noticed that, due to uneven altitude zones specially in the hilly areas, there are some flaws in accuracy found in meteorological parameters of smaller regions within same coordinates. This paper presents a solar powered local weather measurement and prediction system both for plains and hilly regions using Internet of Things (IoT) and Random Forest Regressor. An ESP32 microcontroller, paired with the temperature, humidity and Atmospheric pressure sensors used to facilitate real-time meteorological data monitoring. The proposed system incorporates online weather data for comparison with local-sensor parameters within same coordinates to track out the minor changes. These minor changes feed into the Random-Forest Regressor algorithm for improvement in measurement and prediction accuracy. With 98% calibrated sensor accuracy, this scalable prototype minimizes the errors of local weather parameters of small region which is beneficial to local agriculture practices like terrace cultivation, contour ploughing; meteorological surveys and future researches.</p>

错误：搜索内容不能为空，请输入英文关键词
错误：关键词超出字数限制，请精简
高级检索

Error Detection of Ambient Weather Data under Same Coordinates for Divergent Altitudes in Comparison with Web Data Using Random-Forest Regressor and Internet of Things

  • 期刊论文

  • 发表日期：

    2026年4月9日

查看全文
  • Chiradeep Ghosh,
  • Atanu Chowdhury,
  • Debapam Saha,
  • Himadri Sekhar Dutta

摘要

Online weather forecast system usually works over a large geographical area. However, it has been noticed that, due to uneven altitude zones specially in the hilly areas, there are some flaws in accuracy found in meteorological parameters of smaller regions within same coordinates. This paper presents a solar powered local weather measurement and prediction system both for plains and hilly regions using Internet of Things (IoT) and Random Forest Regressor. An ESP32 microcontroller, paired with the temperature, humidity and Atmospheric pressure sensors used to facilitate real-time meteorological data monitoring. The proposed system incorporates online weather data for comparison with local-sensor parameters within same coordinates to track out the minor changes. These minor changes feed into the Random-Forest Regressor algorithm for improvement in measurement and prediction accuracy. With 98% calibrated sensor accuracy, this scalable prototype minimizes the errors of local weather parameters of small region which is beneficial to local agriculture practices like terrace cultivation, contour ploughing; meteorological surveys and future researches.