About the project
High power fibre lasers (HPFLs), developed first at the University of Southampton, have advanced beyond recognition. Output powers have increased by more than four orders of magnitude in the past two decades, reaching 10kW in a single beam. They are widely used in the most advanced production lines for cutting, welding, 3D printing and marking of a myriad of materials from glass to steel. However, we are now close to the maximum power that can be produced by a single fibre laser.
To continue increasing the power, new solutions must be found. Just as modern computers contain large numbers of processor cores rather than a single high-speed core, the future for HPFLs is in the combination of multiple fibre lasers.
The successful combination of large numbers of fibre lasers would transform manufacturing. Such a breakthrough could enable control of all light properties, such as:
- wavelength
- polarization
- intensity
- phase
This would enable us to create dynamically reconfigurable structured light that changes on the fly depending on the specific application. Such a digital fibre laser would not only make the UK a more prosperous nation, but also allow us to
- protect against malevolent drones
- build the next generation of efficient and compact particle accelerators,
- clean-up space debris
- treat nuclear waste
All-in-all make the world a better, cleaner, greener, and safer place.
The University of Southampton has recently been awarded £6million to solve the challenges associated with the creation of the digital fibre laser, and you will be part of this team effort.
You will be solving the scientific challenges associated with the coherent combination of multiple fibre lasers. This will involve developing a multi-channel master oscillator power amplifier, for on-demand power scaling and controllable beam shaping.
You will then take advantage of the latest development in the field of deep learning to unlock automatic control and optimisation of this fibre laser design. Here, the application of deep learning will be key, as it can provide real-time and sophisticated control and feedback loops, which will be perfect for real-time optimisation of the phase and polarisation of each fibre.
You will have the opportunity to develop experimental expertise in fibre lasers and learn how to apply deep learning techniques to solve a range of extremely complex scientific challenges.
