Additive Manufacturing (AM) is revolutionising UK industry. It is a tool-less digital approach that produces highly customised parts on demand, anywhere in the product life cycle (from prototyping to maintenance and repair) (Gupta, Weber, & Newsome, 2012). Relative to conventional methods, AM can drastically improve component performance, reduce whole lifecycle waste (Frazier, 2014) and create parts with superior mechanical properties (Lott et al., 2011). It facilitates freedom in design and open innovation (Gupta et al., 2012).
We must de-risk AM technology to maximise its impact. AM could transform the healthcare (custom implants, prosthetics, drug delivery), aerospace (lightweight optimised components) (Gupta et al., 2012) and automotive (Giffi, Gangula, & Illinda, 2014) sectors. These disciplines are, however, highly regulated and sensitive to failure. Uncertainties associated with the quality, reproducibility (Lott et al., 2011) and material properties (Mani et al., 2015) of AM parts inhibit significant adoption in these areas (Tapia & Elwany, 2014). This is compounded by high machine-to-machine variability (Frazier, 2014) and difficulties correcting manufacturing errors (which may occur internally to the part). All these issues stem from a lack of process control in Additive Manufacturing.
Machine learning can, potentially, exploit the vast amounts of process measurements that can be captured during an AM build (temperature, back-reflected light etc.) to develop process control strategies. Such an approach, however, involves the analysis of large sets of data. In the current project, each Renishaw AM machine being considered generates approximately 1TB of data per day, while a typical facility consists of around 50 such machines.
Building on previous work at the University of Liverpool (Okaro, Jayasinghe, Sutcliffe, & Black, 2018), the challenge for the current project is to develop scalable machine learning solutions that de-risk AM part development. These solutions must be applicable to large data sets that are generated from different builds, from networks of multiple AM machines.
The project will be conducted with Renishaw, the UK’s only vendor of Laser Powder Bed Fusion (L-PBF) machines. It will investigate how distributed computing architectures can help the storage and analysis of process measurements that are generated from multiple Renishaw L-PBF machines. With regard to machine learning, the project will aim to exploit flexible approaches such as (Gal, Wilk, & Rasmussen, 2014), for example, where Gaussian Processes regression was shown to be applied in a Map-Reduce setting.
This project is part of the EPSRC Funded CDT in Distributed Algorithms: The What, How and where of Next-Generation Data Science. https://www.liverpool.ac.uk/research/research-at-liverpool/research-themes/digital/cdt-distributed-algorithms/
The University of Liverpool is working in partnership with the STFC Hartree Centre and other industrial partners from the manufacturing, defence and security sectors to provide a 4 year innovative PhD training course that will equip over 60 students with the essential skills needed to become future leaders in data science, be it in academia or industry.
Every project within the centre is offered in collaboration with an Industrial partner who as well as providing co-supervision will also offer the unique opportunity for students to access state of the art computing platforms, work on real world problems, benchmarking and data. Our graduates will gain unparalleled experiences working across academic disciplines in highly sought-after topic areas, answering industry need.
As well as learning from academic and industrial world leaders, the centre has a dedicated programme of interdisciplinary research training including the opportunity to undertake modules at the global pinnacle of Data science teaching. A large number of events and training sessions are undertaken as a cohort of PhD students, allowing you to build personal and professional relationships that we hope will lead to research collaboration either now or in your future.
The learning nurtured at this centre will be based upon anticipation of the hardware recourses arriving on desks of students after they graduate, rather than the hardware available today.
To apply for this Studentship please submit an application for an Engineering PhD via our online platform (https://www.liverpool.ac.uk/study/postgraduate-research/how-to-apply/
) and provide the studentship title and supervisor details when prompted. Should you wish to apply for more than one project, please provide a ranked list of those you are interested in.
For a full list of the entry criteria and a recruitment timeline (including interview dates etc), Please see our website https://www.liverpool.ac.uk/research/research-at-liverpool/research-themes/digital/cdt-distributed-algorithms/
Frazier, W. E. (2014). Metal additive manufacturing: A review. Journal of Materials Engineering and Performance, 23(6), 1917–1928. https://doi.org/10.1007/s11665-014-0958-z
Gal, Y., Wilk, M. van der, & Rasmussen, C. (2014). Distributed Variational Inference in Sparse Gaussian Process Regression and Latent Variable Models. Proc. Neural Information Processing Systems (NIPS), 1–9.
Giffi, C. A., Gangula, B., & Illinda, P. (2014). 3D opportunity for the automotive industry. Deloitte University Press, 28. Retrieved from http://dupress.com/articles/additive-manufacturing-3d-opportunity-in-automotive/
Gupta, N., Weber, C., & Newsome, S. (2012). Additive Manufacturing: Status and Opportunities, (March).
Lott, P., Schleifenbaum, H., Meiners, W., Wissenbach, K., Hinke, C., & Bültmann, J. (2011). Design of an optical system for the in situ process monitoring of Selective Laser Melting (SLM). Physics Procedia, 12(PART 1), 683–690. https://doi.org/10.1016/j.phpro.2011.03.085
Mani, M., Lane, B., Donmez, A., Feng, S., Moylan, S., & Fesperman, R. (2015). Measurement Science Needs for Real-time Control of Additive Manufacturing Powder Bed Fusion Processes. https://doi.org/10.6028/NIST.IR.8036
Okaro, I. A., Jayasinghe, S., Sutcliffe, C., & Black, K. (2018). Automatic Fault Detection for Selective Laser Melting using Semi-Supervised Machine Learning, Additive Manufacturing, 10.1016/j.addma.2019.01.006
Tapia, G., & Elwany, A. (2014). A Review on Process Monitoring and Control in Metal-Based Additive Manufacturing. Journal of Manufacturing Science and Engineering, 136(6), 060801. https://doi.org/10.1115/1.4028540
Dr. Xiaohu Guo - STFC Hartree Centre
I’ve been interested in science in my whole life, especially computational science, and can see that it is turning into more and more powerful tool for scientists to explore exciting new areas and unknowns. What excites me about computational science is the great sense of achievement I feel when the algorithms/methods that were turned into codes by my logic and own creativity can be used by every one!
I’m currently a technical lead at the Science and Technology Facilities Council (STFC) Hartree Centre in the development of computational numerical methods using particle method based and unstructured mesh based computation for a range of application fields including computational fluid dynamics (CFD), materials science and medical image processing and reconstruction. What I love about working here is bridging a gap between research academics and industrial applications.
I develop enabling technologies for a wide range of engineering and science applications. I was the lead developer of Incompressible Smoothed Particle Hydrodynamics (ISPH) software package ISPH3D which has been recognised the first open source ISPH software package in the world, which has wide application in the area of nuclear thermal hydraulics, offshore and marine energy industries, offshore oil and gas industries and coastal engineering consultancies involved in the design of coastal defences.
One of my commitments at the moment is to develop the automation and acceleration technologies of digital twinning framework using novel deep learning methods. We’re trying to answer the question of how to find the optimum parameters combination with minimum human intervention.
As an active member of the HPC research community, I am a regular member of technical programme committee for SC and ISC. I have also routinely organised the SIAM minisymposia, in the area of particle method based computational science which have involved major particle application researchers/developers across the world. I am also Specialist Editor in HPC, Grid and Novel Computing, Computer Physics Communications and PhD examiner of Heriot-Watt University.