Gaseous parameters, including species concentration, temperature and velocity are key indicators of the flow-field state and hence play a crucial role in characterizing the thermochemical performance of modern carbon-free power generation systems and advanced manufacturing processes. In many industrial applications, these parameters require to be measured online with high precision and rapidness, to effectively model and active control of the dynamic thermochemical processes.
Laser absorption spectroscopy (LAS) is a non-intrusive optical sensing technology for simultaneous measurement of gas concentration, temperature and velocity. The high-speed tunability of the semi-conductor laser diodes enables the rapid measurement up to MHz. However, complex signal processing of the spectroscopic data is still an issue limiting fast and online measurement. To date, efforts have been made by using machine learning algorithms to accelerate the spectroscopic signal processing. However, most rely on (a) lab-based experimental data that fails to consider the industrial relevant conditions and (b) complicated neural network models that are impractical to be deployed on light computing units.
This project will develop hardware-accelerated laser absorption spectroscopy technology and apply it to industrial gas measurement. The student who joins our group will learn the fundamentals of laser spectroscopic sensing and hardware acceleration techniques, and will be trained for carrying out experiments in both labs and real industrial conditions. The student will have a high chance of working with renowned international researchers and industrial collaborators.
Primary objectives:
1. Develop critical thinking of industrial relevant software and hardware implementation.
2. Develop self-supervised or unsupervised neural networks to process spectroscopic signals sampled under industrial conditions.
3. Deploy the developed neural network on light-weight edge computing platforms.
4. Carry out lab-based and industrial experiments to validate the developments.
5. Improve communication and writing skills via conference presentations and journal publications.
Required skills:
1. Matlab/Python coding (mandatory)
2. Time/frequency-domain signal processing (mandatory)
3. Verilog coding and testing
4. Machine learning algorithms and coding
5. optical/electronic experimental skills.
Please note that this advert will close as soon as a suitable candidate is found.
Further Information:
The University of Edinburgh is committed to equality of opportunity for all its staff and students, and promotes a culture of inclusivity. Please see details here: https://www.ed.ac.uk/equality-diversity
Closing Date:
Principal Supervisor:
Assistant Supervisor:
TBC
Eligibility:
Minimum entry qualification - an Honours degree at 2:1 or above (or International equivalent) in Electronical and Electrical Engineering or Mechanical Engineering, possibly supported by an MSc Degree.
Further information on English language requirements for EU/Overseas applicants.
Funding:
Applications are welcomed from self-funded students, or students who are applying for scholarships from the University of Edinburgh or elsewhere
Informal Enquiries:
Dr Chang Liu, c.liu@ed.ac.uk
