Multi-Domain Airflow Modeling and Ventilation Characterization Using Mobile Robots, Stationary Sensors and Machine...

Area of Science:

• Engineering
• Environmental Science
• Robotics

Background:

• Ventilation systems are vital for safety and health in public buildings and workspaces.
• Current methods for ventilation characterization and high-resolution airflow modeling are limited.
• Accurate airflow data is essential for preventing accidents and mitigating health risks from pollutants.

Purpose of the Study:

• To address the challenge of micro-scale airflow characterization and modeling.
• To develop a novel data-driven algorithm for multi-domain airflow modeling.
• To create high-resolution models of wind speed and direction for ventilation monitoring and robot navigation.

Main Methods:

• A heterogeneous measurement system combining static sensors and mobile robot data collection.
• A novel, data-driven, multi-domain airflow modeling algorithm.
• Estimation of posterior distributions for wind speed/direction (ventilation maps), temporal airflow evolution (ventilation calendars), and frequency domain analysis.

Main Results:

• The proposed algorithm accurately models airflow, even with turbulence and disturbances.
• Validation with simulated data and experiments in semi-controlled and real-world industrial environments.
• Qualitative feedback from plant operators confirmed the models' accuracy and utility in understanding pollutant spread.

Conclusions:

• The developed system provides accurate, high-resolution airflow models for ventilation characterization.
• These models are valuable for ensuring workplace safety, optimizing robot navigation, and understanding pollutant dispersal.
• The approach offers a practical solution for monitoring and maintaining critical ventilation systems.