Register now After registration you will be able to apply for this opportunity online.
Data-Driven Modelling of High-Speed Electronic Nose Response in a PLIF Wind Tunnel
Are you interested in applying data science, signal processing, and machine learning to cutting-edge sensor technology? This project offers an exciting opportunity to work on a unique dataset collected from an advanced wind tunnel experiment, where a high-speed electronic nose (e-nose) was used alongside a Planar Laser-Induced Fluorescence (PLIF) system to monitor gas dispersion. Your task will be to analyse the data, develop models to characterise the e-nose response, and explore methods to extract meaningful information such as response time, transfer function, and saturation effects. The findings will contribute to a better understanding of electronic nose performance in dynamic environments, with applications in environmental monitoring, robotics, and industrial sensing.
Keywords: electronic nose, gas sensors, PLIF, planar laser-induced fluorescence, signal processing, time-series analysis, transfer function, data-driven modelling, machine learning, filtering, predictive modelling, Kalman filters, environmental monitoring, robotics, experimental data analysis
## **What You’ll Work On**
- Processing and analysing multi-sensor time-series data from a high-speed e-nose and PLIF system
- Developing and fitting filters (FIR, IIR, Kalman, ARMAX, etc) or data-driven models (e.g., regression models, machine learning, or system identification techniques) to characterise the e-nose response
- Extracting key performance metrics such as response time, saturation behaviour, and cross-sensitivity
- Investigating the feasibility of real-time predictive filtering based on historical sensor data
- Visualising and interpreting results to identify key insights for future sensor optimisation
## **Who We’re Looking For**
- **BSc/MSc student** in *Electrical Engineering, Mechanical Engineering, Data Science, or a related field*
- Strong background in **signal processing, time-series analysis, or system identification**
- Experience in **Python** for data analysis (*Pandas, NumPy, SciPy, scikit-learn, etc.*)
- Good knowledge of **filtering techniques (Fourier analysis, FIR / IIR / Kalman filters, time series machine learning regression models, etc.)**
- Interest in **sensor technology, gas dynamics, and experimental data analysis**
- Ability to **work independently** and contribute to a high-impact research project
This is a rare opportunity to work with an **unprecedented dataset**—to our knowledge, no study has previously combined high-speed e-nose recordings with PLIF imaging of gas dispersion at this level of detail. The insights gained from this project could lead to significant advancements in sensor modelling, with applications in **environmental monitoring, industrial safety, and robotics**. Additionally, this work is expected to contribute to **high-impact scientific publications**, providing a valuable opportunity for students looking to engage in cutting-edge research.
## **What You’ll Work On**
- Processing and analysing multi-sensor time-series data from a high-speed e-nose and PLIF system - Developing and fitting filters (FIR, IIR, Kalman, ARMAX, etc) or data-driven models (e.g., regression models, machine learning, or system identification techniques) to characterise the e-nose response - Extracting key performance metrics such as response time, saturation behaviour, and cross-sensitivity - Investigating the feasibility of real-time predictive filtering based on historical sensor data - Visualising and interpreting results to identify key insights for future sensor optimisation
## **Who We’re Looking For**
- **BSc/MSc student** in *Electrical Engineering, Mechanical Engineering, Data Science, or a related field* - Strong background in **signal processing, time-series analysis, or system identification** - Experience in **Python** for data analysis (*Pandas, NumPy, SciPy, scikit-learn, etc.*) - Good knowledge of **filtering techniques (Fourier analysis, FIR / IIR / Kalman filters, time series machine learning regression models, etc.)** - Interest in **sensor technology, gas dynamics, and experimental data analysis** - Ability to **work independently** and contribute to a high-impact research project
This is a rare opportunity to work with an **unprecedented dataset**—to our knowledge, no study has previously combined high-speed e-nose recordings with PLIF imaging of gas dispersion at this level of detail. The insights gained from this project could lead to significant advancements in sensor modelling, with applications in **environmental monitoring, industrial safety, and robotics**. Additionally, this work is expected to contribute to **high-impact scientific publications**, providing a valuable opportunity for students looking to engage in cutting-edge research.
To=be-defined together
To=be-defined together
Nik Dennler (dennlern@ethz.ch)
Please use "STUDENT PROJECT: [Title]" as the header in your email.
Nik Dennler (dennlern@ethz.ch)
Please use "STUDENT PROJECT: [Title]" as the header in your email.
