This book introduces an innovative approach to multi-fidelity probabilistic optimisation for aircraft composite structures, addressing the challenge of balancing reliability with computational cost. Probabilistic optimisation pursues statistically reliable and robust solutions by accounting for uncertainties in data, such as material properties and geometry tolerances. Traditional approaches using high-fidelity models, though accurate, are computationally expensive and time-consuming, especially when using complex methods such as Monte Carlo simulations and gradient calculations.
For the first time, the proposed multi-fidelity method combines high- and low-fidelity models, enabling high-fidelity models to focus on specific areas of the design space, while low-fidelity models explore the entire space. Machine learning technologies, such as artificial neural networks and nonlinear autoregressive Gaussian processes, fill information gaps between different fidelity models, enhancing model accuracy. The multi-fidelity probabilistic optimisation framework is demonstrated through the reliability-based and robust design problems of aircraft composite structures under a thermo-mechanical environment, showing acceptable accuracy and reductions in computational time.
Contents:
- Introduction
- Fundamentals of Structural Optimisation
- Multi-Fidelity Models
- Multi-Fidelity Reliability-Based Design Optimisation
- Multi-Fidelity Robust Design Optimisation Using Successive High-Fidelity Correction
- Multi-Fidelity Probabilistic Optimisation Using Sparse High-Fidelity Information
- Conclusion
Readership: This book targets undergraduate and postgraduate students in the fields of aerospace engineering, mechanical engineering, and design engineering. It is also aimed at professional engineers and researchers in the aircraft, motor, civil engineering, wind energy, offshore oil & gas, and naval architecture industries.
"The authors tackle a key challenge in aerospace engineering: creating lighter, more efficient, and sustainable aircraft without compromising safety. By simplifying the complexities of traditional probabilistic design, they offer a framework that allows early consideration of more variables — essential for addressing large-scale design problems."
Dimitrios Bekas
Airbus Operations GmbH
"[The book] not only provides rigorous theoretical foundations but also illustrates practical applications across engineering systems and data science problems ... It is a valuable resource for anyone looking to expand their expertise in multi-fidelity optimization or seeking innovative methodologies to address uncertainty in complex systems."
Zahra Sharif Khodaei
Imperial College London
"The book's clear explanations and practical examples make it an excellent resource for educators and students alike. Its blend of theoretical insights and applied case studies provides instructors with a strong foundation for teaching complex topics in an engaging and accessible way."
Llewellyn Morse
University College London
Dr Kwangkyu Alex Yoo is a Senior Machine Learning Scientist at Deep.Meta in London, a Google-funded AI start-up. Alex completed his PhD in Multi-Fidelity Probabilistic Optimisation for Composite Structures at Imperial College London in 2021. Following his PhD, Alex served as a Research Associate in Industrial Machine Learning at the University of Cambridge. Prior to his doctoral studies, he was a Research Engineer specialising in offshore engineering at Hanwha Ocean in Korea. Alex has consistently developed machine learning and optimisation algorithms to tackle technical challenges across design, manufacturing, operations and supply chains. His expertise lies in machine learning, multi-fidelity modelling, and optimisation under uncertainty, making him a versatile contributor to both academic and industrial advancements. His work is widely recognised through journal articles, conference papers, patents, invited talks, and commercial projects.
Dr Omar Bacarreza is a Senior Machine Learning Scientist at ORCA Computing, working on quantum computing applied to machine learning and optimisation. He has a PhD in Computational Mechanics from the Czech Technical University in Prague. He has more than 30 publications in optimisation, composite materials, and structural health monitoring. He has participated in several projects involving academic and industrial partners. He has used high-performance computing in computational mechanics and machine learning-assisted stochastic design and optimisation of aircraft components. He also has experience with AutoML techniques for improving and finding better deep learning models.
Professor M H Ferri Aliabadi is a Professor of Aerostructures at Imperial College London. He has worked in the field of Solids and Structures and has established an international reputation for his achievements in the development of Computational Methods related to Fracture & Damage Mechanics. Prof. Aliabadi has pioneered a new generation of boundary element methods and is noted for his contributions to other fields. During the last decade, he has pursued research and development in Structural Health Monitoring (SHM) for composite airframes in collaboration with the aeronautics industry (Airbus, Leonardo) supported by significant EU funding. He has been PI and coordinator of several CleanSky EU projects including SMASH and SCOPE and coordinated the SHM platform for the wing in the SARISTU project. His latest project is CleanSky II, a core-partnership with SHERLOC in which he is the PI and coordinator, seeking to develop the next generation of smart (highly sensorised) composite airframes. Prof. Aliabadi has published 500 papers in leading international journals and 65 books related to Experimental and Computational Methods in Solids and Structures.
