Data-Driven Surrogate-Assisted Evolutionary Fluid Dynamic Optimisation

Computational fluid dynamics (CFD) is fundamental to modern engineering design, from aircraft and cars to household appliances. It allows the behaviour of fluids to be computationally simulated and new designs to be evaluated. Finding the best design is nonetheless very challenging because of the vast number of designs that might be explored. Computational optimisation is a crucial technique for modern science, commerce and industry. It allows the parameters of a computational model to be automatically adjusted to maximise some benefit and can reveal truly innovative solutions. For example, the shape of an aircraft might be optimised to maximise the computed lift/drag ratio.

A very successful suite of methods to tackle optimisation problems are known as evolutionary algorithms, so-called because they are inspired by the way evolutionary mechanisms in nature optimise the fitness of organisms. These algorithms work by iteratively proposing new solutions (shapes of the aircraft) for evaluation based upon recombinations and/or variations of previously evaluated solutions and, by retaining good solutions and discarding poorly performing solutions, a population of optimised solutions is evolved.

An obstacle to the use of evolutionary algorithms on very complex problems with many parameters arises if each evaluation of a new solution takes a long time, possibly hours or days as is often the case with complex CFD simulations. The great number of solutions (typically several thousands) that must be evaluated in the course of an evolutionary optimisation renders the whole optimisation infeasible. This research aims to accelerate the optimisation process by substituting computationally simpler, dynamically generated "surrogate" models in place of full CFD evaluation. The challenge is to automatically learn appropriate surrogates from a relatively few well-chosen full evaluations. Our work aims to bridge the gap between the surrogate models that work well when there are only a few design parameters to be optimised, but which fail for large industry-sized problems.

Our approach has several inter-related aspects. An attractive, but challenging, avenue is to speed up the computational model. The key here is that many of these models are iterative, repeating the same process over and over again until an accurate result is obtained. We will investigate exploiting partial information in the early iterations to predict the accurate result and also the use of rough early results in place of the accurate one for the evolutionary search. The other main thrust of this research is to use advanced machine learning methods to learn from the full evaluations how the design parameters relate to the objectives being evaluated. Here we will tackle the computational difficulties associated with many design parameters by investigating new machine learning methods to discover which of the many parameters are the relevant at any stage of the optimisation. Related to this is the development of "active learning" methods in which the surrogate model itself chooses which are the most informative solutions for full evaluation. A synergistic approach to integrate the use of partial information, advanced machine learning and active learning will be created to tackle large-scale optimisations.

An important component of the work is our close collaboration with partners engaged in real-world CFD. We will work with the UK Aerospace Technology Institute and QinetiQ on complex aerodynamic optimisation, with Hydro International on cyclone separation and with Ricardo on diesel particle tracking. This diverse range of collaborations will ensure research is driven by realistic industrial problems and builds on existing industrial experience. The successful outcome of this work will be new surrogate-assisted evolutionary algorithms which are proven to speed up the optimisation of full-scale industrial CFD problems.

Aerospace and Aviation

The UK has the second largest aerospace industry in the world with significant capabilities in the key areas, including engines, airframe structures, aerodynamics and air traffic. To maintain the UK's world-leading position in this sector, optimisation techniques like evolutionary algorithms will provide a unique opportunity to address the main complex challenges and create innovative designs. This research will remove the obstacles in applying evolutionary techniques in the aerospace and aviation sector.

One of our project partners, the Aerospace Technology Institute (ATI), was created by the government in 2012 as part of the wider aerospace strategy. As their letter of support attests, the work is "clearly aligned with the key challenges faced in trying to use optimisation for future high-lift aerodynamic design". We will work closely with QinetiQ, a member of the ATI, on the CFD optimisation in complex, nonlinear flows, work which directly addresses the ATI's national strategy for aerospace technology in the UK.



Industry

In addition to aerospace and aviation, computational fluid dynamics is used across a wide spectrum of UK industry: in civil engineering (e.g., water systems engineering), environmental engineering (e.g., flood prevention; wind turbines; carbon capture systems), mechanical engineering (e.g., automotive design), biomedical engineering (e.g., anaerobic digesters), the food industry and medical devices. In all these arenas optimisation is used to optimise a benefit, such as efficiency, safety, manufacturing or deployment cost, quality of service, energy consumption or health. Being able to effectively optimise these systems therefore clearly has direct and wide socio-economic benefits. The successful outcome of this project will result in new, efficient ways of optimising large, complex systems. These are systems that, because of their complexity, cannot currently be optimised and are therefore operated sub-optimally. The potential impact of this project is therefore very broad. Moreover the UK industrial CFD community is a significant one with industrial research and development of CFD ongoing as well as application through various specialist consultancies.

Optimisation is increasingly used throughout UK industry and commerce, as well as the public sector and throughout science. In addition to the directly CFD-related applications, the breadth of its use is indicated by applications in air traffic safety, radio and mobile telephone networks, drug design, supply chain configuration, renewable energy devices, energy and transport networks, additive layer manufacturing, automotive structures and so on. We therefore anticipate that the methods developed for evolutionary surrogate-assisted optimisation will have impact across a wide range of the UK economy. We will release open source code to facilitate widespread use of the work.


Public impact

The project will have public impact through Exeter and Surrey's public engagement and outreach programmes, such as F1 in Schools, Future Engineers and Formula Student. It will also impact school workshops interested in computer science and engineering.


Education and Training

This project will train three post-doctoral research associates and an Exeter-funded PhD student to work in at the interface between computational fluid dynamics, evolutionary computing and machine learning. In addition, by spending significant amounts of time working with our industrial partners, they will be able to understand and tackle real-world problems. It will also have an impact on the undergraduate Engineering and Computer Science degree programmes in Surrey and Exeter, through seminars and case studies, involvement in undergraduate teaching and final year projects. The project will therefore have impact by delivering highly skilled people to UK industry and commerce.