A Multi-Parameter Measurement System for MEMS Anemoters for Data Collection with Machine Learning Outcomes

Thomas Hackett*, Dennis Alveringh, Remco G.P. Sanders, Tom E. van den Berg, Jurriaan Schmitz

*Corresponding author for this work

Research output: Contribution to conferencePaper

2 Downloads (Pure)

Abstract

In order to generate consistent and comprehensive datasets for the application of
machine learning algorithms to MEMS thermal flow sensors, a measurement set up was created.This system allows automatic data collection of large datasets involving parameters such as the angle of attack, humidity, temperature and flow speed. The electrical output signals in both the time and frequency domain can be measured for both AC and DC actuation. The setup has been able to fully characterize an anemometer by exposing it to flows of 0 to 5 m/s in steps of 0.02 m/s under angles
from -45 to 45° in steps of 5° at a constant temperature of 25 °C and humidity of 30 %RH and complete the measurement in 8 hours.
Original languageEnglish
Pages126-129
Number of pages4
Publication statusPublished - 22 Feb 2024
Event5th Conference on MicroFluidic Handling Systems, MFHS 2024 - Munich, Germany
Duration: 21 Feb 202423 Feb 2024
Conference number: 5

Conference

Conference5th Conference on MicroFluidic Handling Systems, MFHS 2024
Abbreviated titleMFHS 2024
Country/TerritoryGermany
CityMunich
Period21/02/2423/02/24

Keywords

  • MEMS
  • Machine Learning
  • Thermal
  • Anemometer
  • Microfluidics
  • Flow sensor

Fingerprint

Dive into the research topics of 'A Multi-Parameter Measurement System for MEMS Anemoters for Data Collection with Machine Learning Outcomes'. Together they form a unique fingerprint.

Cite this