A Machine Learning Enhanced MEMS Thermal Anemometer for Detection of Flow, Angle of Attack, and Relative Humidity

Thomas Hackett; Remco G.P. Sanders; Jurriaan Schmitz; Dennis Alveringh; Tom E. van den Berg; Jeong Young Choi

doi:10.1109/LSENS.2024.3418193

A Machine Learning Enhanced MEMS Thermal Anemometer for Detection of Flow, Angle of Attack, and Relative Humidity

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

^*Corresponding author for this work

Research output: Contribution to journal › Article › Academic › peer-review

21 Downloads (Pure)

Abstract

By optimizing machine learning (ML), the accuracy of a thermal anemometer has been improved (511%) when compared to conventional linear regression. In addition, ML has extended the functionality allowing for additional angle of attack and humidity information to be determined. The miniature sensor (0.16 cm 2 ) has been fabricated with a straightforward silicon on insulator (SOI) fabrication procedure. The sensor paired with ML could offer a cost-effective, small, and reliable solution for monitoring air in industrial and agricultural sensor grid applications, such as data centers and greenhouses. This proof of principle shows that thermal anemometers can have their accuracy and functionality enhanced through ML, enabling the estimation of multiple physical parameters with a single sensor.

Original language	English
Article number	6008304
Number of pages	4
Journal	IEEE Sensors Letters
Volume	8
Issue number	7
Early online date	24 Jun 2024
DOIs	https://doi.org/10.1109/LSENS.2024.3418193
Publication status	Published - Jul 2024
Event	IEEE SENSORS 2024 - Kobe Portopia Hotel, Kobe, Japan Duration: 20 Oct 2024 → 23 Oct 2024 https://2024.ieee-sensorsconference.org/

Keywords

Sensor applications
angle measurement
flow sensor
humidity
machine learning (ML)
thermal anemometer

Access to Document

10.1109/LSENS.2024.3418193Licence: CC BY-NC-ND

A_Machine_Learning_Enhanced_MEMS_Thermal_Anemometer_for_Detection_of_Flow_Angle_of_Attack_and_Relative_HumidityFinal published version, 2.36 MBLicence: CC BY-NC-ND

1 Conference contribution
1 Paper

A Multi-Parameter Measurement System for MEMS Anemoters for Data Collection with Machine Learning Outcomes
Hackett, T., Alveringh, D., Sanders, R. G. P., van den Berg, T. E. & Schmitz, J., 22 Feb 2024, p. 126-129. 4 p.
Research output: Contribution to conference › Paper
A miniature microclimate thermal flow sensor for horticultural applications
Alveringh, D., Bijsterveld, D. G., Berg, T. E. V. D., Veltkamp, H.-W., Batenburg, K. M., Sanders, R. G. P., Lötters, J. C. & Wiegerink, R. J., 8 Dec 2022, 2022 IEEE Sensors. IEEE, 4 p. 9967348
Research output: Chapter in Book/Report/Conference proceeding › Conference contribution › Academic › peer-review

Open Access
File
2 Citations (Scopus)

102 Downloads (Pure)

Cite this

@article{cb5349805ed04b3b8c8bac47dd31787c,

title = "A Machine Learning Enhanced MEMS Thermal Anemometer for Detection of Flow, Angle of Attack, and Relative Humidity",

abstract = "By optimizing machine learning (ML), the accuracy of a thermal anemometer has been improved (511%) when compared to conventional linear regression. In addition, ML has extended the functionality allowing for additional angle of attack and humidity information to be determined. The miniature sensor (0.16 cm 2 ) has been fabricated with a straightforward silicon on insulator (SOI) fabrication procedure. The sensor paired with ML could offer a cost-effective, small, and reliable solution for monitoring air in industrial and agricultural sensor grid applications, such as data centers and greenhouses. This proof of principle shows that thermal anemometers can have their accuracy and functionality enhanced through ML, enabling the estimation of multiple physical parameters with a single sensor.",

keywords = "Sensor applications, angle measurement, flow sensor, humidity, machine learning (ML), thermal anemometer",

author = "Thomas Hackett and Sanders, {Remco G.P.} and Jurriaan Schmitz and Dennis Alveringh and {van den Berg}, {Tom E.} and Choi, {Jeong Young}",

year = "2024",

month = jul,

doi = "10.1109/LSENS.2024.3418193",

language = "English",

volume = "8",

journal = "IEEE Sensors Letters",

issn = "2475-1472",

publisher = "IEEE",

number = "7",

note = "IEEE SENSORS 2024 ; Conference date: 20-10-2024 Through 23-10-2024",

url = "https://2024.ieee-sensorsconference.org/",

}

TY - JOUR

T1 - A Machine Learning Enhanced MEMS Thermal Anemometer for Detection of Flow, Angle of Attack, and Relative Humidity

AU - Hackett, Thomas

AU - Sanders, Remco G.P.

AU - Schmitz, Jurriaan

AU - Alveringh, Dennis

AU - van den Berg, Tom E.

AU - Choi, Jeong Young

PY - 2024/7

Y1 - 2024/7

N2 - By optimizing machine learning (ML), the accuracy of a thermal anemometer has been improved (511%) when compared to conventional linear regression. In addition, ML has extended the functionality allowing for additional angle of attack and humidity information to be determined. The miniature sensor (0.16 cm 2 ) has been fabricated with a straightforward silicon on insulator (SOI) fabrication procedure. The sensor paired with ML could offer a cost-effective, small, and reliable solution for monitoring air in industrial and agricultural sensor grid applications, such as data centers and greenhouses. This proof of principle shows that thermal anemometers can have their accuracy and functionality enhanced through ML, enabling the estimation of multiple physical parameters with a single sensor.

AB - By optimizing machine learning (ML), the accuracy of a thermal anemometer has been improved (511%) when compared to conventional linear regression. In addition, ML has extended the functionality allowing for additional angle of attack and humidity information to be determined. The miniature sensor (0.16 cm 2 ) has been fabricated with a straightforward silicon on insulator (SOI) fabrication procedure. The sensor paired with ML could offer a cost-effective, small, and reliable solution for monitoring air in industrial and agricultural sensor grid applications, such as data centers and greenhouses. This proof of principle shows that thermal anemometers can have their accuracy and functionality enhanced through ML, enabling the estimation of multiple physical parameters with a single sensor.

KW - Sensor applications

KW - angle measurement

KW - flow sensor

KW - humidity

KW - machine learning (ML)

KW - thermal anemometer

U2 - 10.1109/LSENS.2024.3418193

DO - 10.1109/LSENS.2024.3418193

M3 - Article

SN - 2475-1472

VL - 8

JO - IEEE Sensors Letters

JF - IEEE Sensors Letters

IS - 7

M1 - 6008304

T2 - IEEE SENSORS 2024

Y2 - 20 October 2024 through 23 October 2024

ER -

A Machine Learning Enhanced MEMS Thermal Anemometer for Detection of Flow, Angle of Attack, and Relative Humidity

Abstract

Keywords

Access to Document

Fingerprint

Research output

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

A miniature microclimate thermal flow sensor for horticultural applications

Cite this