Home MaPS nohref

Fifth Knowledge Transfer Workshop/Sandpit on Inverse Problems in Industry
6 - 7 June 2011
School of Mathematics, University of Leeds

The Fifth Knowledge Transfer Workshop/Sandpit on Inverse Problems in Industry will be held between 6-7 June 2011 at the University of Leeds.
It is intended that the workshop/sandpit will commence on Monday 6th of June around 11.00 and finish on Tuesday 7th of June around 12.30.

The purpose of the meeting is for industrialists and academics to identify inverse problems of common interest.
Inverse Problems seek to identify causes from given effects and as such they are at the heart of many physical phenomena and any advance into the subject should be encouraged.

The Workshop/Sandpit will follow the format of presentations from industry by way of introducing the problems on the first day, followed by intensive work on the problems by groups (academic and industrial) during the next day.

The workshop is supported by the Research Impact & Innovation Fund offered by the University of Leeds.

As an outcome of this activity it is hoped that a rapport between academics and industrialists will be established and cemented through consultancy, possible grant proposals for internships, PhD Studentships / CASE Awards, post-docs, etc. to be submitted to the EPSRC or to other foundations.

Academic participation is open to all members of the inverse problems community or related subjects, including postgraduate students, for whom the proposed activity provides an excellent training in oriented research.

Please note that there are no fees to be charged for participating at the workshop.

To register for the workshop, academics and industrialists are invited to send their contact details (before 4th of May 2011) to:

Professor Daniel Lesnic
Department of Applied Mathematics,
University of Leeds,
Leeds LS2 9JT, UK.
e-mail: amt5ld@maths.leeds.ac.uk,
tel: +44-(0)113-3435181,
fax: +44-(0)113-3435090

The programme of talks from industry includes:

(UK National Nuclear Laboratory (NNL))
"Applied mathematical modelling in the nuclear industry"

ABSTRACT. In this presentation a potential range of applications that can be attacked using mathematical modelling (e.g. CFD, Heat Transfer and Stress, Seismic and Impact Analysis) will be discussed.
Nevertheless, some of these applications will be suitable candidates for inverse problem methodologies and it is hoped that this will widely open the discussion and collaboration.

Larissa FRADKIN and Victor ZERNOV
(Sound Mathematics Ltd.)
"Defect characterisation in ultrasonic non-destructive testing"

ABSTRACT. NDT (Non-Destructive Testing) of industrial components is routinely carried around the world using an ever increasing number of methods and devices. We concentrate on ultrasonic NDT, since this is the least expensive and most widely used approach throughout a range of industries, for example, nuclear, railway and airspace; the applications are growing in oil and gas, chemical and power industries.
At present, defect chracterisation is carried out by highly skilled technicians who are trained to interpret echoes of ultrasonic pulses emitted by ultrasonic transducers or transducer arrays.
as the industries grow and more and more emphasis is put on health monitoring of industrial structures there exists are real need for developing practical inverse methods for automatic or at least semi-automatic defect characterisation.

(Schlumberger Cambridge Research Ltd.)
"Challenges in seismic imaging and inversion"

ABSTRACT. We will give a general overview of some of the challenges involved in modern seismic imaging for hydrocarbon exploration. The talk will be illustrated with a number of examples and we will describe some of the computational challenges which arise. Aspects of nonlinear optimization and the numerical solution of the wave equation will be discussed. The talk will be suitable for a general audience.

(Computational Modelling Cambridge Ltd.)
"Parameter estimation with uncertainty propagation for applications in the chemical, energy, and automotive industries"

ABSTRACT. The demand for robust models which can be used to analyse physical phenomena spans numerous industrial sectors such as chemical, energy, and automotive.
Despite the diversity of these applications, many of the challenges associated with using and applying models in these sectors are shared, particularly in terms of storing data observed from experiments, identifying the most suitable parameters for the model and applying the model with confidence.
CMCL innovations develops software tools which enable engineers to parameterise models using experimental data whilst accounting for the uncertainties in the measurements. Once identified, parameters carry both a mean value as well as an associated uncertainty, and when applied, yield model predictions along with their confidence bounds. This enables users to identify regimes of high and low model robustness; in an effort to improve model performance, engineers can then focus their model development activities on those sub-processes of the model that are characterised by parameters which carry greater uncertainties.
Two applications of these tools are highlighted in this talk: 1) estimation of unknown model parameters for a granulation process (chemical industry application) and 2) a methodology for automatic model development and reduction of corresponding model parameter uncertainties applied to emissions reduction in modern Diesel engines.