Accelerating Scientific Discoveries through AI and Machine Learning
There has been a significant influx of research using artificial intelligence (AI) to find novel methods for simulating engineering materials like metal alloys and composites in recent years. However, this research often neglects simple principles of AI and machine learning, such as an understanding of the underlying data. Typical engineering applications around machine learning suffer from small and fragmented physical datasets (from experiments) with largely under-represented classes.
Dr Johannes believes that a combination of physics-based simulations and machine learning is essential to exploit the full potential of data-driven engineering around AI. He hopes that raising awareness of the typical pitfalls of using AI in mechanical engineering will help to emphasise AI and machine learning’s potential to significantly accelerate scientific discoveries in the field of engineering materials and also transform other branches of science and engineering.
A Career in Composite Materials
Dr Johannes gained his Doctorate in Mechanical Engineering from the University of Queensland in 2016. After 2.5 years as a Postdoctoral Fellow at the University of British Columbia, Canada, he was appointed Lecturer in Mechanical Engineering at Deakin University, Australia in July 2019. He has been involved in a range of industry projects focused on developing AI-driven simulation methods to predict the mechanical behaviour of lightweight composite structures.
Quantifying Composite Behaviour
Dr Johannes’ current DIA-supported project aims to develop a virtual AI-enabled platform with the ability to predict the mechanical response of composite structures made from specific manufacturing processes. This will result in optimising existing manufacturing procedures and designs and exploring new fabrication techniques and material systems.
Dr Johannes recognises the potential AI and machine learning have in transforming how composite structures are understood and improved in manufacturing and engineering. The DIA grant allows him to collaborate with Professor Stephen Hallett and Dr Jonathan Belnoue from the University of Bristol to incorporate AI into the simulation of composite materials while accounting for their mechanical behaviour and the uncertainties associated with them.
Quantifying Composite Behaviour
Dr Johannes’ current DIA-supported project aims to develop a virtual AI-enabled platform with the ability to predict the mechanical response of composite structures made from specific manufacturing processes. This will result in optimising existing manufacturing procedures and designs and exploring new fabrication techniques and material systems.
Dr Johannes recognises the potential AI and machine learning have in transforming how composite structures are understood and improved in manufacturing and engineering. The DIA grant allows him to collaborate with Professor Stephen Hallett and Dr Jonathan Belnoue from the University of Bristol to incorporate AI into the simulation of composite materials while accounting for their mechanical behaviour and the uncertainties associated with them.
The role of a Distinguished International Associate provides me with a great platform to inspire and engage colleagues seeking to explore AI in engineering
Engineering an Inclusive Workforce
Dr Johannes hopes that the AI-enabled platform will create an inclusive workforce to which everyone can contribute. To this end, he is working on developing the AI-based computational framework to allow current and future engineering workforces to be trained promptly. This will help to preserve and expand valuable, long-learned knowledge by establishing AI-driven sovereign manufacturing capabilities in Australia and the UK.
Dr Johannes’ ongoing collaborations with institutions in the aerospace sector also expand the potential of the DIA-supported project being utilised to understand and improve carbon fibre composite and sustainable wood-based composite structures in aerospace applications.