Bristol’s research in dynamical systems methods are adopted and made operational within Airbus UK to develop cost-saving, high-precision modelling platforms
Submitting Institution
University of BristolUnit of Assessment
Mathematical SciencesSummary Impact Type
TechnologicalResearch Subject Area(s)
Mathematical Sciences: Applied Mathematics
Engineering: Electrical and Electronic Engineering
Summary of the impact
Evaluating the ground-based manoeuvrability of large aircraft is time
consuming and costly if explored though industry-developed complete models
of ground dynamics. Research by Krauskopf and colleagues from the
University of Bristol has shown that applying methods from dynamical
systems allow these dynamics to be investigated efficiently and with
considerable precision. This approach, and the related purpose-developed
software, Dynamical Systems Toolbox, have been adopted by Airbus.
It is now fully incorporated in the Airbus Methods and Tools
portfolio as a supported tool for the evaluation of proposed works and new
designs. The research delivers considerable savings in time and costs for
the company. Additionally, this programme of research has delivered
research training for Airbus employees and one, who studied for PhD with
Krauskopf, now leads the Airbus development and implementation of these
mathematical techniques which are being disseminated more widely within
the company. There continue to be Bristol EPSRC CASE PhD studentships in
collaboration with Airbus co-supervised by Krauskopf (7 in the assessment
period).
Underpinning research
Aircraft are designed to fly and, hence, are not optimised for ground
operations in the way a car or other ground vehicles may be. Nevertheless,
a passenger aircraft needs to operate fast, reliably and safely on the
ground in order to ensure its overall commercial success. Prior to this
research, the standard methodology of aircraft manufacturers was to
conduct massive and expensive numerical simulations of industry-tested and
parametrised models for aspects of aircraft motion to evaluate the ground
performance of aircraft as part of their design, evaluation and
certification.
The underpinning research consists of the systematic mathematical
analysis of several of these models using dynamical systems techniques to
study their nonlinear dynamics and the stabilities of particular aspects
of their motion. This methodology avoids expensive brute-force numerical
simulations. The models studied represent different features of ground
manoeuvres and were developed in collaboration with Airbus. The behaviour
of these nonlinear mathematical models was evaluated using both analytical
and numerical techniques and their parameter dependence was analysed using
software, purpose-developed by Krauskopf and colleagues, the Dynamical
Systems Toolbox, which runs under MATLAB (a widely used numerical
computing environment and programming language). This enabled the
introduction of numerical continuation to follow solutions and their
changes of stability with varying parameters. Importantly, this approach
could be used within an industrial setting.
Specific projects were as follows (references from section 3 in
parentheses):
- a fully parameterized model of a mid-size aircraft was used to
evaluate aircraft ground turning and provide guidelines for the safe
operation limits and maximum load factors under different conditions.
The main motivation for this work was to evaluate the suitability of the
existing Federal Aviation Regulations for lateral loads experienced
during turning maneuvers [3].
- bifurcation studies of shimmy oscillations of an aircraft nose landing
gear. A mathematical model with torsional and lateral bending modes that
are coupled through a wheel-mounted elastic tyre was developed and
studied. The bifurcation analysis was performed in terms of the forward
velocity & vertical force and was used to find regions of stable
torsional and stable lateral shimmy oscillations [1].
- research on ground handling studies of aircraft with more than three
sets of wheels, in particular, the Airbus A380 model. This led to the
development of the Dynamical Systems Toolbox and demonstrated
its usefulness with the evaluation of low-, medium- and high-speed
ground manoeuvers of an A380 in comparison with an A320 [2,5,6].
- treatment of the dynamics of a deployment cycle of an aircraft landing
gear, showing the usefulness of dynamical systems methods with a study
of actuator placement for nose and main landing gears [4].
The various strands of this research programme were undertaken in Bristol
from 2006 onwards and have led to a series of scientific publications in
which the underlying mathematics, and its application to aircraft
dynamics, are explained. The publications appeared in high quality applied
mathematics journals. The research was led by Krauskopf (University of
Bristol until 2011, University of Auckland from 2011), Lowenberg
(University of Bristol) and Neild (University of Bristol), together with
Bristol postgraduate students and postdoctoral research assistants Rankin,
Knowles, Thota and Desroches, and industrial contacts and supervisors at
Airbus [b,c]. There are currently three EPSRC Case students in Bristol
still co-supervised by Krauskopf.
References to the research
*[3] Rankin, J., Krauskopf, B., Lowenberg, M. and Coetzee, E. 2010, Nonlinear
analysis of lateral loading during taxiway turns, Journal
of Guidance, Control, and Dynamics (American Institute of
Aeronautics and Astronautics) 33(6) 1708-1717;
DOI:10.2514/1.50626
* references that best indicate the quality of the underpinning research.
Details of the impact
The introduction of dynamical systems methods, which are implemented in
the Dynamical Systems Toolbox, into the industrial practice at
Airbus provides the company with new capability for the evaluation of
aircraft ground performance and the potential of considerable savings in
time and costs (estimated at 80% of staff time compared to current
practice within the aircraft industry [c]). Airbus has fully adopted this
methodology and has developed a strategy for the continued future use of
dynamical systems methods in an internal document, Strategic Research
into Dynamical Systems Methods: A roadmap from 2009 to 2013 [a].
The software, Dynamical Systems Toolbox (DST), has transitioned
within Airbus from its Research portfolio into its Methods and Tools
portfolio. This is a mandatory stage of Airbus for institutionalisation of
new operational engineering methods. Further evidence of the integration
of these techniques into Airbus is the development of the Dynamical
Systems Toolbox interface to Airbus' own software tools was
sub-contracted and internal training sessions have been held. The new tool
is presently in use by members of the Future Project Office, the Process,
Methods & Tools Department and the Landing Gear group; the latter work
on current aircraft projects, as well as the exploration of new landing
gear and aircraft concepts. It is estimated that the new tool has made
significant savings in terms of man hours during 2012[c]. Airbus UK have
described the impact of the research as approximately 80% reduction in
time and associated costs [c] for ground maneuverability analysis and
specified that the estimate of the saving is made by comparing the time
taken to conduct global assessments using bifurcation and other dynamical
systems techniques versus the time to obtain exhaustively a large number
of point solutions. This is important because it allows the efficient use
of numerical simulations for new designs. The engineer from the Modelling
& Simulations Methods and Tools group writes, "The tool provides a
method for de-risking design-decisions; the trade-off being between the
impact of a risk occurring versus the likelihood of the risk occurring"
[b].
The research formed the basis of the enhanced capability introduced by
Airbus Process, Methods & Tools via the development of fully
parameterized and validated models, the development of the Dynamical
Systems Toolbox and the demonstration of the approach via dedicated
case studies. The research has been supported and sponsored by Airbus, via
direct interaction between the industry and university based researchers.
The research has grown considerably in reaction to demand from the company
— now also including novel testing strategies and the characterization of
aircraft upset scenarios in flight. Regular research meetings (about every
six months) were conducted at University of Bristol and at Airbus at which
further dissemination could occur, together with the identification of new
projects. Additionally, there have been several subsequent PhD students
associated with this research project who have each spent extended
placements at the company. Evidence of the value of this project to Airbus
comes in part from the large number of grants and support they have
provided (£415K in direct funding, plus funding in kind) [d].
There is also impact from the work in terms of staff training in research
and recruitment in Airbus. For example, a member of staff was seconded
from Airbus to perform PhD research at the University of Bristol. He is
now one of the main advocates of the use of dynamical systems methods and
is helping to lead their introduction within Airbus. In addition, a former
postdoctoral research assistant (held 2007-2010) at Bristol was hired by
Airbus and now works as Airbus Model Developer — Physical Systems at
Airbus' Design Analysis.
A further impact of this research has been the embedding of nonlinear
dynamics approaches within Airbus. It is expected that usage of the DST
will increase amongst other departments when research through currently
on-going PhD projects near completion (there are several on-going Bristol
EPSRC Mathematics CASE studentships in partnership with Airbus). There are
plans for further training courses within Airbus and improvements in the
DST interface. Airbus also plans to set up a more extensive training
course with the University of Bristol.
Sources to corroborate the impact
[a] Airbus internal report Strategic Research into Dynamical Systems
Methods: A roadmap from 2009 to 2013. Details from this report are
available from a Future Projects Engineer, Airbus Operations Ltd.
Corroborates contribution of methodology developed at the University of
Bristol to Airbus's strategy for the continued future use of dynamical
systems methods.
[b] Expert (Modelling & Simulation Methods and Tools, Airbus
Operations Ltd. Can be contacted to corroborate the contribution of
the research to Airbus' decision making around risk.
[c] Future Projects Engineer, Airbus Operations Ltd.
Can be contacted to corroborate claim of 80% reduction in time and cost
for Airbus.
[d] This research activity received a total of £805K of funding, of which
£415K from industrial partner Airbus, as follows:
(i) PI: Krauskopf, Co-I: Lowenberg. Title: Bifurcation analysis of
nonlinear ground handling of aircraft. Sponsor: EPSRC Maths Case Award
with Airbus in the UK. Period: October 2006 - March 2010 to support PG
Rankin. Value: £65K (EPSRC) plus £25K (Airbus)
(ii) PI: Krauskopf, Co-I: Lowenberg. Title: Analysis of nonlinear ground
handling models. Sponsor: Airbus in the UK. Period: April 2007 - March
2010 to support PDRA Thota. Value: £220K
(iii) PI: Krauskopf, Co-I: Lowenberg. Title: Modelling and nonlinear
analysis of landing gear and airframe interactions. Sponsor: Airbus in the
UK. Period: March 2007 - June 2011 to support PhD work of Coetzee (while
full-time employee of Airbus). Value: salary cost for Coetzee plus £25K
(research support)
(iv) PI: Krauskopf, Co-I: Lowenberg. Title: Dynamics of statically
indeterminate landing gear systems. Sponsor: EPSRC Maths Case Award with
Airbus in the UK. Period: October 2008 - March 2012 to support PG Knowles.
Value: £65K (EPSRC) plus £25K (Airbus)
(v) PI: Krauskopf, Co-I: Lowenberg. Title: Dynamics of aircraft main
landing gears. Sponsor: EPSRC Maths Case Award with Airbus in the UK.
Period: October 2009 - March 2013 to support PG Howcroft. Value: £65K
(EPSRC) plus £30K (Airbus)
(vi) PI: Neild, Co-I: Krauskopf. Title: Dynamic substructure testing
strategies in aerospace. Sponsor: EPSRC Industrial Case Award with Airbus
in the UK. Period: August 2010 - January 2014 to support PG Terkovics.
Value: £65K (EPSRC) plus £30K (Airbus)
(vii) PI: Lowenberg, Co-I: Krauskopf. Title: Investigation of airliner
upset and upset recovery dynamics using bifurcation analysis. Sponsor:
EPSRC Maths Case Award with Airbus in France. Period: October 2010 - March
2014 to support PG Gill. Value: £65K (EPSRC) plus £30K (Airbus)
(viii) PI: Krauskopf, Co-I: Lowenberg, Neild. Title: Investigation of
coupled landing gear and fuselage vibrations. Sponsor: EPSRC Maths Case
Award with Airbus in the UK. Period: October 2011 - March 2015 to support
PG Kewley. Value: £65K (EPSRC) plus £30K (Airbus).
Corroborates Airbus's investment in underpinning research.