Skip to main content
Go to the home page of the European Commission (opens in new window)
English en
CORDIS - EU research results
CORDIS

A Universal Acoustic MEMS Gas Sensor with Machine Learning

Project description

That gas ‘sounds’ familiar: an acoustic gas sensor harnessing microresonator technology

Sensors detect or measure physical properties in many ways and in a virtually limitless number of applications. When it comes to sensing molecules in a gas, most sensors currently rely on the detection of electrical or optical signals. The EU-funded SmartGas project plans to expand options with an innovative acoustic gas sensor harnessing microelectromechanical systems (MEMS) technology, acoustic resonance and machine learning. The acoustic resonance patterns of a gas or gases inside the tiny cavity of a MEMS resonator will replace the optical frequencies in spectroscopy. The project plans to deliver a patented design for a low-cost, universal gas sensor that could revolutionise the gas sensing market.

Objective

I am proposing a novel acoustic MEMS gas sensor with machine learning for the first time that can revolutionize the gas sensing field. Gas sensors targeting only a specific type of gas are developed due to the nature of the existing sensing technologies. Current gas sensors either rely on the detection of the electrical property changes upon the reaction of an active material with the gas or infrared (IR) transmission/absorption characteristics of the gases. I am proposing to couple the acoustic resonance in a cavity with a MEMS resonator and use the coupled resonance and damping as a gas sensor. Acoustic frequency and the damping are a function of the gas inside the cavity, and together can be used to detect any gas with machine learning. Since the frequency and damping change with multiple gases in the proposed approach, machine learning algorithms can extract the gas changes in a smart way. Proposed sensor solves the problems of the current gas sensors. It can be manufactured with standard MEMS fabrication flows, can be used with any gas, does not saturate, and has immediate response time. The sensors will be fabricated in the fully equipped clean room UNAM in Bilkent University. I will apply the sensor on human health by working with a pulmonologist to measure CO in human breath and indoor air quality. My supervisor Prof. Hilmi Volkan Demir will guide me throughout the project and help me to improve my technical and soft skills. I will also be able to improve my network and knowledge during my secondment in Fraunhofer EMFT, Germany. Collaborating with Analog Devices and Fraunhofer will be the key for further product technology development. In summary, this fellowship will establish me as a recognized European leader by demonstrating a proof of concept with interdisciplinary research (MEMS, machine learning, and medical application), by patenting and publishing my innovation, and by enabling the path for low cost universal gas sensor productization.

Fields of science (EuroSciVoc)

Keywords

Programme(s)

Topic(s)

Funding Scheme

MSCA-IF - Marie Skłodowska-Curie Individual Fellowships (IF)

See all projects funded under this funding scheme

Call for proposal

(opens in new window) H2020-WF-2018-2020

See all projects funded under this call

Coordinator

BILKENT UNIVERSITESI VAKIF
Net EU contribution
€ 157 355,52
Address
ESKISEHIR YOLU 8 KM
06800 Bilkent Ankara
Türkiye
Region
Batı Anadolu Ankara Ankara
Activity type
Higher or Secondary Education Establishments
Links
Total cost
€ 157 355,52