Scholarlink

The main goal of this paper is to present the design and implementation of an Internet of Things (IoT)–based environmental monitoring and control system that integrates real-time data acquisition, database management, machine learning (ML), and intelligent decision-making for real time monitoring and prediction of environmental parameters. The proposed system collects temperature, humidity, and light intensity data using sensors from the Elegoo Basic Kit connected to an Elegoo Uno R3 microcontroller. The data are stored in an Oracle 23c Free database (freepdb1) and processed in MATLAB and Python to build predictive models. Linear regression models are trained to forecast environmental parameters and to automate control decisions such as turning lights on/off or adjusting temperature and humidity through a PID controller. A Streamlit-based web interface allows users to visualize real-time and predicted values, compare MATLAB and Python model performance, and observe control actions. Based on these predictions, the system automatically adjusts controls, as toggling lights, activating air conditioning or heating, and running humidifiers/dehumidifiers. Experimental results demonstrate the feasibility of integrating IoT data acquisition with baseline ML prediction and automated control actions, while also highlighting the limitations of simple linear models for highly dynamic real-world signals. The proposed system is applicable to small-scale building automation and educational experimentation.

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

IoT-Based Environmental Monitoring and Control System Using Machine Learning Models in MATLAB and Python

  • 会议论文

  • 发表日期：

    2026年

查看全文
  • Gentian Qinami,
  • Elinda Kajo Meçe,
  • Aida Spahiu

摘要

The main goal of this paper is to present the design and implementation of an Internet of Things (IoT)–based environmental monitoring and control system that integrates real-time data acquisition, database management, machine learning (ML), and intelligent decision-making for real time monitoring and prediction of environmental parameters. The proposed system collects temperature, humidity, and light intensity data using sensors from the Elegoo Basic Kit connected to an Elegoo Uno R3 microcontroller. The data are stored in an Oracle 23c Free database (freepdb1) and processed in MATLAB and Python to build predictive models. Linear regression models are trained to forecast environmental parameters and to automate control decisions such as turning lights on/off or adjusting temperature and humidity through a PID controller. A Streamlit-based web interface allows users to visualize real-time and predicted values, compare MATLAB and Python model performance, and observe control actions. Based on these predictions, the system automatically adjusts controls, as toggling lights, activating air conditioning or heating, and running humidifiers/dehumidifiers. Experimental results demonstrate the feasibility of integrating IoT data acquisition with baseline ML prediction and automated control actions, while also highlighting the limitations of simple linear models for highly dynamic real-world signals. The proposed system is applicable to small-scale building automation and educational experimentation.